AN INTRODUCTION

TO PHONETICS

Parviz Birjandi (PhD)

Allameh Tabatabaii University

Mohammad Ali Salmani-Nodoushan (PhD)

University of Zanjan

ZABANKADEH PUBLICATIONS
8 Bazarcheh Ketaab, Enghelab Avenue, Tehran, Iran
Phone: +98 21 66402367
Fax: +98 21 66492961
© 2005 by Zabankadeh Publications
No 8, Bazarcheh Ketaab, Enghelab Sq., P.O. Box: 13145-564 Tehran, Iran
E-Mail: info@zabankadeh.net Tel.: 02166402367 Fax: 02166492961
All rights reserved. No part of this book may be reproduced in any form, by photostat,
microfilm, xerography, or any other means, or incorporated into any information retrieval
system, electronic or mechanical, without permission in writing from the copyright owner.
An Introduction to Phonetics
Authors: Parviz Birjandi (PhD) Mohammad Ali Salmani-Nodoushan (PhD)
Printed in Iran
Salmani-Nodoushan, Mohammad Ali ‫ ـ‬1348 ،‫ ﻣﺤﻤﺪﻋﻠﻲ‬،‫ﺳﻠﻤﺎﻧﻲ ـ ﻧُﺪوﺷﻦ‬
(‫)ان اﻳﻨﺘﺮوداﻛﺸﻦ ﺗﻮ ﻓُﻮﻧِﺘﻴﻜﺲ‬
An Introduction to Phonetics / Mohammad Ali Salmani-Nodoushan, Parviz Birjandi.‫ـ‬
.‫ م‬2005 = 1384 ، ‫ زﺑﺎﻧﻜﺪه‬: ‫ﺗﻬﺮان‬
.‫ ص‬180 ، xii
ISBN: 964-6117-57-0
.‫اﻧﮕﻠﻴﺴﻲ‬
.‫ﻓﻬﺮﺳﺘﻨﻮﻳﺴﻲ ﺑﺮ اﺳﺎس اﻃﻼﻋﺎت ﻓﻴﭙﺎ‬
.180 ‫ ـ‬177 .‫ ص‬:‫ﻛﺘﺎﺑﻨﺎﻣﻪ‬
.‫ زﺑﺎن اﻧﮕﻠﻴﺴﻲ ـ ـ واج ﺷﻨﺎﺳﻲ‬.2 .‫ زﺑﺎن اﻧﮕﻠﻴﺴﻲ ـ ـ آواﺷﻨﺎﺳﻲ‬.1
.An Introduction to Phonetics :‫ ﻋﻨﻮان‬.‫ ب‬.Birjandi, Parviz ، -1319 ،‫ ﭘﺮوﻳﺰ‬،‫ ﺑﻴﺮﺟﻨﺪي‬.‫اﻟﻒ‬
421 /5 PE 1135 / ‫س‬8‫اﻟﻒ‬9
1384
‫م‬84 ‫ ـ‬6550 ‫ﻛﺘﺎﺑﺨﺎﻧﻪ ﻣﻠﻲ اﻳﺮان‬

An Introduction to Phonetics
‫ دﻛﺘﺮ ﻣﺤﻤﺪﻋﻠﻲ ﺳﻠﻤﺎﻧﻲ ﻧُﺪوﺷﻦ‬/ ‫ دﻛﺘﺮ ﭘﺮوﻳﺰ ﺑﻴﺮﺟﻨﺪي‬:‫ﻣﻮﻟﻔﺎن‬
‫اﻧﺘﺸﺎرات زﺑﺎﻧﻜﺪه‬
‫ ﭼﺎپ دﻳﺒﺎ‬،‫ ﻧﺴﺨﻪ‬3000 ‫ ﺗﻴﺮاژ‬،1384 ‫ﭼﺎپ اول‬
‫ ﺗﺮﺟﻤﻪ ﻣﺘﻦ و ﻫﺮ اﺳﺘﻔﺎده‬،‫ ﻫﺮﮔﻮﻧﻪ ﻧﺴﺨﻪ ﺑﺮداري‬.‫ﻛﻠﻴﻪ ﺣﻘﻮق ﺑﺮاي اﻧﺘﺸﺎرات زﺑﺎﻧﻜﺪه ﻣﺤﻔﻮظ ﻣﻲ ﺑﺎﺷﺪ‬
.‫دﻳﮕﺮ از ﻣﺘﻦ ﻛﺘﺎب ﻣﻤﻨﻮع ﺑﻮده و ﻣﺘﺨﻠﻒ ﺗﺤﺖ ﭘﻴﮕﺮد ﻗﺎﻧﻮﻧﻲ ﻗﺮار ﺧﻮاﻫﺪ ﮔﺮﻓﺖ‬

8 ‫ ﺷﻤﺎره‬،‫ ﺑﺎزارﭼﻪ ﻛﺘﺎب‬،‫ روﺑﺮوي دﺑﻴﺮﺧﺎﻧﻪ داﻧﺸﮕﺎه ﺗﻬﺮان‬،‫ﺗﻬﺮان‬

(021) 66492961 :‫( ﻓﺎﻛﺲ‬021) 66402367 :‫ﺗﻠﻔﻦ‬
‫ رﻳﺎل‬23000 :‫ﻗﻴﻤﺖ‬ 13145 ‫ ـ‬564 :‫ﺻﻨﺪوق ﭘﺴﺘﻲ‬
ISBN: 964-6117-57-0 964-6117-57-0 :‫ﺷﺎﺑﻚ‬
 ACKNOWLEDGMENTS

Many people have played an important role in the development of this book.
Colleague Richard W. Sorfleet, of Ontario College of Teachers, Canada, was
extremely supportive of this book, offering numerous suggestions for our
consideration. He read the manuscript and provided us with useful comments
which enabled us to convert turgid prose into readable English.

Special acknowledgment and thanks are also extended to colleagues Dr Mehdi

Nowroozi Khiabani and Dr Kurosh Lachini for their insightful reviews of the
manuscript. Their comments and suggestions were very valuable to us. Our
special thanks also go to Mrs. Arezou Pashapur for typing the manuscript. We
would also like to thank the people at Zabankadeh Publications for publishing
and marketing the book.

Parviz Birjandi
Mohammad Ali Salmani-Nodoushan
August 2005

III
 CONTENTS

PREFACE XI

CHAPTER ONE: PRELIMINARIES

1. INTRODUCTION 1
2. EARLY CONTRIBUTORS 1
3. THE ROLE OF PHILOSOPHY 2
4. THE ROLE OF PHYSIOLOGY AND PSYCHOLOGY 3
5. THE ROLE OF AMERICAN STRUCTURALISM 4
6. THE EMERGENCE OF PHONETICS 6
EXERCISE 7

CHAPTER TWO: ARTICULATORY PHONETICS: RUDIMENTS

1. INTRODUCTION 9
2. PHONEME 9
3. THE NEED FOR PHONETIC WRITING SYSTEM 10
4. BROAD AND NARROW TRANSCRIPTION 16
5. SYLLABLE STRUCTURE 19
6. RECEIVED PRONUNCIATION 22
EXERCISE 23

CHAPTER THREE: TRADITIONAL ARTICULATORY PHONETICS

1. INTRODUCTION 27
2. SPEECH ARTICULATION 32
2.1. MANNER OF ARTICULATION 32
2.1.1. OCCLUSIVE AND PLOSIVE 33
2.1.2. FRICATIVE 35
2.1.3. LATERAL 40

V
 2.1.4. APPROXIMANT 40
2.1.5. NASAL 43
2.1.6. AFFRICATE 44
2.1.7. VOICING (LEVEL OF VIBRATION) 45
2.1.8. FORTIS AND LENIS 47
2.1.9. LIP ROPUNDING 47
2.2. PLACE OF ARTICULATION 47
2.2.1. LIPS 48
2.2.2. TEETH 48
2.2.3. ALVEOLAR RIDGE 49
2.2.4. HARD PALATE 50
2.2.5. THE SOFT PALATE (VELUM) 50
2.2.6. GLOTTIS 50
3. CONSONANTS 51
4. VOWELS 55
4.1. THE VOWEL CHART 61
4.2. TYPOLOGY OF VOWELS 62
4.3. WEAK FORMS OF VOWELS 64
5. SEMIVOWELS OR GLIDES 66
6. LIQUIDS 67
EXERCISE 67

CHAPTER FOUR: SYSTEMATIC ARTICULATORY PHONETICS

1. INTRODUCTION 71
2. WHAT IS SYSTEMATIC/MODERN PHONETICS 71
3. AIRSTREAM MECHANISMS 79
4. PHONETIC FEATURES 80
4.1. VOCALIC VERSUS CONSONANTAL 80
4.2. VOICE 81
4.3. NASAL AND NASALIZATION 82
4.4. ABRUPT RELEASE 84

VI
 4.5. CONTINUANT 85
4.6. ASPIRATION 86
4.7. ANTERIOR 87
4.8. CORONAL 89
4.9. STRIDENT 90
4.10. LATERAL 91
4.11. SONORANT 92
4.12. SYLLABIC 93
EXERCISE 95

CHAPTER FIVE: SUPRASEGMENTALS

1. INTRODUCTION 99
2. SUPRASEGMENTALS 99
2.1. STRESS 99
2.1.1. THE PLACE OF STRESS 101
2.1.2. STRESS VERSUS PROMINENCE 102
2.1.3. THE IMPORTANCE OF STRESS 103
2.1.3.1. BOUNDARY MARKING 104
2.1.3.2. ADDITIONAL CONTRASTS 104
2.1.4. RULES OF STRESS 105
2.1.4.1. CONTENT AND FUNCTION WORDS 106
2.1.4.2. WORD STRESS PATTERNS 106
2.1.4.2.1. MONOSYLLABIC WORDS 107
2.1.4.2.2. BI-SYLLABIC WORDS 108
2.1.4.2.3. MULTI-SYLLABIC WORDS 109
2.2. INTONATION 113
2.2.1. WHAT INTONATION DOES 114
2.2.1.1. EXPRESSING NEW INFORMATION 114
2.2.1.2. CONTRAST 115
2.2.1.3. MEANING 116
2.2.1.4. PRONUNCIATION 116

VII
 2.2.1.5. MOOD AND PERSONALITY 117
2.2.2. TYPES OF INTONATION 118
2.2.2.1. FALLING INTONATION 119
2.2.2.2. RISING INTONATION 120
2.2.2.3. RISING-FALLING INTONATION 122
2.2.2.4. FALLING-RISING INTONATION 123
2.2.2.5. TAKE-OFF INTONATION 124
2.2.2.6. LEVEL INTONATION 124
2.2.3. PHRASING VERSUS INTONATION 125
2.3. TONE 125
2.4. LENGTH 126
2.5. SYLLABLE 126
EXERCISE 127

CHAPTER SIX: PHONEMICS

1. INTRODUCTION 129
2. SEQUENTIAL/PHONOLOGICAL CONSTRAINTS 130
3. PHONOLOGICAL PROCESSES 131
3.1. ASSIMILATION 131
3.2. ELISION 134
3.3. LIAISONS (JOINING) 137
3.3.1. CONSONANT-VOWEL SEQUENCES 139
3.3.2. CONSONANT-CONSONANT SEQUENCES 139
3.3.3. VOWEL-VOWEL SEQUENCES 140
3.3.4. /T, D, S, Z/+/Y/ SEQUENCES 140
3.4. NEUTRALIZATION 142
3.5. VOICE ONSET TIME (VOT) 143
3.6. ASPIRATION 144
3.7. NASALIZATION 145
3.8. NON-CONTIGUOUS ASSIMILATION 146
EXERCISE 146

VIII
CHAPTER SEVEN: ACOUSTIC PHONETICS
1. INTRODUCTION 149
2. WAVE MOTION 150
3. TYPES OF WAVES 151
4. TYPES OF WAVE MOTION 152
5. SOUND 153
6. PHYSICAL CHARACTERISTICS 154
6.1. FREQUENCY 155
6.2. AMPLITUDE 157
6.3. INTENSITY 158
6.4. QUALITY 158
7. SPEED OF SOUND 160
8. REFRACTION, REFLECTION, AND INTERFERENCE 161
9. TYPES OF ORDINARY SOUND 162
10. SPEECH SIGNAL PROCESSING 163
EXERCISE 163

CHAPTER EIGHT: AUDITORY PHONETICS

1. INTRODUCTION 165
2. STRUCTURE OF THE HUMAN EAR 165
2.1. OUTER EAR 166
2.2. MIDDLE EAR 167
2.3. INNER EAR 168
3. HEARING 168
4. SPEECH PERCEPTION 172
5. SENSATIONS OF TONE 175
EXERCISE 176

BIBLIOGRAPHY 177

IX
 PREFACE

A major difficulty facing almost any foreign language learner is the achievement
of acceptable pronunciation which marks his success in mastering the
language. Many EFL learners master such aspects of language as syntax,
semantics, morphology, and even pragmatics to the point of native-like
competence, but fail to master phonology. This is partly because of the
physiological constraints that make the pronunciation of a foreign or second
language sound different from that of the native language of the speakers, and
partly due to the lack of appropriate training in phonology courses.

An Introduction to phonetics is designed to support EFL learners in achieving

native-like pronunciation:

• Chapter one deals with the history of phonology and phonetics and
provides a brief overview of the impact of philosophy and psychology
on the emergence of phonology.

• Chapter two defines the notion of phoneme, describes IPA phonetic

alphabet, and distinguishes between broad and narrow transcriptions.

• Chapters three and four provide an in-depth account of traditional and

systematic articulatory phonetics respectively.

• Chapter five discusses the place of suprasegmentals in phonology.

• Chapter six seeks to explain phonemics.

• Chapter seven provides a brief introduction to the rudiments of acoustic

or physical phonetics.

• Chapter eight introduces the reader to the notion of auditory phonetics.

XI
 • The tables, figures, and photos that are presented throughout the book
are designed to give the reader an instant reference for the precise
articulation of English phonemes.

The book is designed for use in undergraduate classes of phonology and

phonetics. The fifth chapter is also useful for students of conversation classes.
Teachers at high school level may also find the fifth chapter valuable. The step
by step approach of the book towards its subject matter makes it is easy for the
reader to follow the line of discussion without the help of a phonology teacher.

Parviz Birjandi
Mohammad Ali Salmani-Nodoushan
August 2005

XII
 CHAPTER ONE
PRELIMINARIES

1. INTRODUCTION

Phonetics is a branch of linguistics which is concerned with the production,

physical nature, and perception of speech sounds. The main fields of study are
experimental phonetics, articulatory phonetics, phonemics, acoustic phonetics,
and auditory phonetics.

2. EARLY CONTRIBUTORS

The earliest contributions to phonetics were made more than 2000 years ago
by Sanskrit scholars such as the grammarian Panini in the 400s who dealt with
articulation to keep the pronunciation of ancient rituals unchanged. The first
phonetician of the modern world was Dane J. Matthias, author of De Litteris
(1586). English mathematician John Wallis, who instructed deaf-mutes, was
the first to classify vowels, in 1653, according to their place of articulation. The
vowel triangle was invented in 1781 by C. F. Hellwag from Germany. Ten years
later, Austrian mechanician Wolfgang von Kempelen invented a machine that
produced speech sounds. German physicist Hermann Helmholtz, who wrote
Sensations of Tone (1863), inaugurated the study of acoustic phonetics.
Frenchman Abbé Jean Pierre Rousselot pioneered in experimental phonetics.

Late in the 19th century, the theory of the phoneme was advanced by Jan
Baudouin de Courtenay from Poland and Ferdinand de Saussure from
Switzerland. In the United States, linguist Leonard Bloomfield and
anthropologist and linguist Edward Sapir contributed greatly to the phonetic
theory. Linguist Roman Jakobson developed a theory of the universal
characteristics of all phonemic systems. Perhaps Ferdinand de Saussure

CHAPTER ONE 1
(1857-1913), a Swiss linguist, has provided the clearest ideas about phonetics.
His book Memoir on the Original Vowel System in the Indo-European
Languages (1879) is an important work on the vowel system of Proto-Indo-
European—considered the parent language from which the Indo-European
languages descended.

3. THE ROLE OF PHILOSOPHY

Recent scientific endeavors all get their insights from the developments in
philosophy. In fact, schools of philosophy have shed light on the paths in which
different branches of science have stepped. Phonetics, like other fields, has
been deeply influenced—at least in its origins—by philosophy. Specifically, the
ideas of Kant have had the greatest impact on the early steps of phonetics.

Perhaps the greatest influence is due to the Scottish philosopher and historian
David Hume who argued that no observable evidence is available for the
existence of a mind substance, spirit, or God. In other words, all metaphysical
assertions about things that cannot be directly perceived (that is, through the
five sense modalities of sight, taste, touch, hearing, and/or smell) are equally
meaningless and non-existent, he claimed, and should be “committed to the
flames.” Following Hume's line of thought, Immanuel Kant later categorized
knowledge to two major branches: (a) the “phenomenal world” of experience,
and the noumenal world of faith. He maintained that metaphysical beliefs about
the soul, the cosmos, and God are matters of faith rather than of scientific
knowledge.

Along the same lines, early linguists, and most notably among them Leonard
Bloomfield maintained that language only consists of elements which are
perceivable through the five sense modalities. That is, the only scientifically
acceptable parts of language are those elements which we can experience
through our five senses. As such, meaning was out of consideration in linguistic
theories. The most readily tangible part of language was its phonological part.
Therefore, linguists were expected to focus on the study of sounds.

2 CHAPTER ONE
4. THE ROLE OF PHYSIOLOGY AND PSYCHOLOGY

The origins of phonetics are also traceable in physiology and psychology.

Perhaps the most famous physiologist whose ideas were widely incorporated
into the developments of phonetics was Ivan Pavlov. He is noted for his
pioneer work in the physiology of the heart, nervous system, and digestive
system. His most famous experiments, begun in 1889, demonstrated the
conditioned and unconditioned reflexes in dogs, and they had an influence on
the development of physiologically oriented behaviorist theories of psychology
during the early years of the 20th century. His work on the physiology of the
digestive glands won him the 1904 Nobel Prize in physiology or medicine.

Pavlov carried out a series of experiments in which he provided what he called

stimuli for his dog and measured what he called responses. In these
experiments, Pavlov accompanied the call of a bell with dog-food and
measured the secretion of saliva by the dog. He repeated this stimulus several
times. When he predicted that the dog could associate the ring of the bell with
the presence of food, Pavlov omitted the food but kept the bell. To his surprise,
Pavlov observed that the dog's secretion of saliva in the absence of food
increased as if food was present. That is, the dog had learnt to react to the call
of the bell. Pavlov concluded that organisms could learn new behavior upon the
repetition of appropriate stimuli to the point of automaticity. Pavlov's
experiments led to the notion of Behaviorism in psychology which is, in turn, at
the heart of structuralism in linguistics.

Behaviorism is a movement in psychology that advocates the use of strict

experimental procedures to study observable behavior (or responses) in
relation to the environment (or stimuli). The behavioristic view of psychology
has its roots in the writings of the British associationist philosophers, as well as
in the American functionalist school of psychology and the Darwinian theory of
evolution, both of which emphasize the way that individuals adapt and adjust to
the environment.

CHAPTER ONE 3
Behaviorism was first developed in the early 20th century by the American
psychologist John B. Watson. The dominant view of that time was that
psychology is the study of inner experiences or feelings by subjective,
introspective methods. Following Kantian philosophy, Watson did not deny the
existence of inner experiences, but he insisted that these experiences could
not be studied because they were not observable. He was greatly influenced by
the pioneering investigations of the Russian physiologists Ivan Pavlov and
Vladimir Bekhterev on conditioning of animals (i.e., classical conditioning).
Watson proposed to make the study of psychology scientific by using only
objective procedures such as laboratory experiments designed to establish
statistically significant results. The behavioristic view led him to formulate a
stimulus-response (S-R) theory of psychology. In this theory all complex forms
of behavior—emotions, habits, and such—are seen as composed of simple
muscular and glandular elements that can be observed and measured. He
claimed that emotional reactions are learned in much the same way as other
skills.

Watson's stimulus-response theory resulted in a tremendous increase in

research activity on learning in animals and in humans, from infancy to early
adulthood. Between 1920 and midcentury, behaviorism dominated psychology
in the United States and also had wide international influence. By the 1950s,
the new behavioral movement had produced a mass of data on learning that
led such American experimental psychologists as Edward C. Tolman, Clark L.
Hull, and B. F. Skinner to formulate their own theories of learning and behavior
based on laboratory experiments instead of introspective observations.

5. THE ROLE OF AMERICAN STRUCTURALISM

Early linguistic studies were concerned with historical issues such as language
families. As linguists in the United States became involved in the study of
American Indian languages during the late 19th and early 20th centuries, it
became increasingly clear that the historical orientation of nineteenth-century

4 CHAPTER ONE
European linguistics was not very practical for work with languages that lacked
extensive written materials from the past. Furthermore, many American
linguists of this period were also anthropologists—scientists who study human
beings.

These factors led to a speech-oriented focus in the linguistic studies in the

United States. Scholars turned their attention to the form of languages,
emphasizing the description of the phonological and other surface forms of
these languages. American linguists frequently approached American Indian
languages without the advantage of knowing even the basic sound system, let
alone the principles of sentence formation or meaning of these languages.

Thus, they had to start their studies with what was most immediately
observable in these languages—the sounds. Since few linguists ever achieved
this degree of fluency in American Indian languages, American linguistics of the
first half of the twentieth century was characterized by an intensive
investigation of sounds and the principles of word formation. Little (if at all)
attention was paid to syntax (i.e., word-order) or semantics (i.e., meaning).

Because of its attention to the form, or structure of language, American

linguistics of the early 20th century came to be known as Structural Linguistics.
Attempting to describe languages that they themselves could not speak,
American linguists were forced to concentrate on the directly observable
aspects of these languages—their sounds. In their work, they gradually evolved
a set of procedures considered useful in determining the sound system of
languages. These procedures and techniques were later known as Discovery
Procedures.

Despite the limitations of this early work on American Indian languages, the
detailed, objective investigations of linguists during the first half of the twentieth
century provided concrete evidence about the diversity that exists among
human languages.

CHAPTER ONE 5
6. THE EMERGENCE OF PHONETICS

All these developments resulted in the emergence of a new branch within

linguistics which was concerned with the study of the phonological component
of language—phonology. Phonology is the study of all aspects of the sounds
and sound system of a language. It includes two major sub-branches: (a)
phonetics, and (b) phonemics.

Phonetics is the field of language study concerned with the physical properties
of sounds, and it has three subfields:

(a) articulatory phonetics (i.e., the study of how the human vocal organ
produces sound)
(b) acoustic phonetics (i.e., the study of the sound waves produced by the
human vocal apparatus)
(c) auditory phonetics (i.e., the examination of how speech sounds are
perceived by the human ear)

Phonemics, in contrast, is not concerned with the physical properties of

sounds. Rather, it focuses on how sounds function in a particular language.
The following example illustrates the difference between phonetics and
phonology. In the English language, when the sound /k/ (usually spelled c)
occurs at the beginning of a word, as in the word cut, it is pronounced with
aspiration (a puff of breath). However, when this sound occurs at the end of a
word, as in tuck, there is no aspiration. Phonetically, the aspirated [kh] and
unaspirated [k] are different sounds, but in English these different sounds never
distinguish one word from another or bring about differences in meaning, and
English speakers are usually unaware of the phonetic difference until it is
pointed out to them. Thus English makes no phonological distinction between
the aspirated and unaspirated /k/. The Hindi language, however, uses this
sound difference to distinguish words such as kal (meaning time), which has an
unaspirated /k/, and khal (meaning skin), in which /kh/ represents the aspirated
[kh]. Therefore, in Hindi the distinction between the aspirated and unaspirated

6 CHAPTER ONE
/k/ is both phonetic and phonological—phonemic (i.e., any difference in
pronunciations which brings about a difference in meaning is said to be
phonemic). The following chapters will provide an in-depth discussion of both
phonetics (i.e., auditory, acoustic, and articulatory phonetics) and phonemics.

EXERCISE

Decide whether the following statements are true or false. Mark Ⓕ for false

and Ⓣ for true statements.

01 Phonetics is concerned with the production, physical nature, and Ⓣ Ⓕ

perception of speech sounds.

02 The main fields of phonetics and phonology are experimental Ⓣ Ⓕ

phonetics, articulatory phonetics, acoustic phonetics, auditory
phonetics, and phonemics.

03 The theory of the universal characteristics of all phonemic Ⓣ Ⓕ

systems was developed by Roman Jacobson.

04 Because of its emphasis on the five senses, Kantian philosophy Ⓣ Ⓕ

helped the emergence of phonetics.

05 Phonology is the study of all aspects of the sounds and sound Ⓣ Ⓕ

system of a language.

06 The major difference between phonetics and phonology is that Ⓣ Ⓕ

phonetics is concerned with the description of speech sounds
while phonology is concerned with the description of the sound
system of language.

07 The branch of phonetics that has to do with the study of how the Ⓣ Ⓕ
human vocal organ produces speech sounds is called auditory
phonetics.

CHAPTER ONE 7
08 Phonetics and phonemics are the sub-branches of phonology. Ⓣ Ⓕ

09 Phonemics focuses on how sounds function in a particular Ⓣ Ⓕ

language.

10 American structuralism is the school of linguistics which had the Ⓣ Ⓕ

most influence on the emergence and development of phonology.

8 CHAPTER ONE
 CHAPTER TWO
ARTICULATORY PHONETICS: RUDIMENTS

1. INTRODUCTION
Articulatory phonetics describes speech sounds genetically—that is, with
respect to the ways by which the vocal organs modify the air stream in the
mouth, nose, and throat in order to produce a sound. To date, articulatory
phonetics has witnessed two major movements: (a) traditional phonetics, and
(b) modern or systematic phonetics. These movements will be discussed in
detail in the following chapters. This chapter will focus on the rudimentary
topics which are crucial to the understanding of articulatory phonetics.

To show how a speech sound is articulated, all the vocal activities involved in
the production of a sound need not be described. Only a selection of them such
as the place and manner of articulation is enough. Phonetic symbols and their
articulatory definitions are abbreviated descriptions of these selected activities.
The symbols most commonly used are those adopted by the International
Phonetic Association (IPA) and are written in brackets or between slant lines.
An understanding of articulatory phonetics requires an understanding of a
number of basic concepts usually employed by phoneticians in their discussion
of phonetics.

2. PHONEME

The basic building block of any discussion of articulatory phonetics is

phoneme. The technical term phoneme is usually used to refer to sound
segments. Linguists define phoneme as the minimal unit of sound (or
sometimes syntax). The study of phonemes is the study of the sounds of
speech in their primary function, which is to make vocal signs that refer to the
fact that different things sound different. The phonemes of a particular

CHAPTER TWO 9
language are those minimal distinct units of sound that can distinguish meaning
in that language. In English, the /p/ sound is a phoneme because it is the
smallest unit of sound that can make a difference of meaning if, for example, it
replaces the initial sound of such words as bill, till, or dill, making the word pill.
The vowel sound /ɪ/ of pill is also a phoneme because its distinctness in sound
makes pill, which means one thing, sound different from pal, which means
another. Two different sounds, reflecting distinct articulatory activities, may
represent two phonemes in one language but only a single phoneme in
another. Thus phonetic /r/ and /l/ are distinct phonemes in English, whereas
these sounds represent a single phoneme in Japanese, just as [ph] and [p] in
pie and spy, respectively, represent a single phoneme in English although
these sounds are phonetically distinct.

Phonemes are not letters; they refer to the sound of a spoken utterance. For
example, flocks and phlox have exactly the same five phonemes. Similarly, bill
and Bill are identical phonemically, regardless of the difference in meaning.
Each language has its own inventory of phonetic differences that it treats as
phonemic (that is, as necessary to distinguish meaning). For practical
purposes, the total number of phonemes for a language is the least number of
different symbols adequate to make an unambiguous graphic representation of
its speech that any native speaker could read if given a sound value for each
symbol, and that any foreigner could pronounce correctly if given additional
rules covering nondistinctive phonetic variations that the native speaker makes
automatically. For convenience, each phoneme of any language may be given
a symbol.

3. THE NEED FOR PHONETIC WRITING SYSTEM

It is quite clear that ordinary writing systems (i.e., alphabet or orthography)

cannot illustrate pronunciation differences. In fact, there are many words that
are written with the same set of letters but pronounced differently. There are
also some words which are written with different sets of letters but pronounced

10 CHAPTER TWO
the same. As such, phonologists and phoneticians felt the need for a new
writing system, one in which the symbols fully represent the sounds of any
language.
Standard English orthography (the writing system) cannot capture all the
sounds of English pronunciation. The same is true for writing systems of many
of the world's languages. Even if the writing system does a good job of
capturing all the sounds of a language, what happens when the pronunciation
of a word changes over time? Or when there are multiple pronunciations for a
word? To overcome this, linguists use the phonetic alphabet, designed to
represent all the possible sounds of the world's languages in a standard way.
The most commonly-used alphabet is known as the Phonetic Alphabet
designed by the International Phonetic Association (IPA) in the late 19th
century.

The anomaly between writing and pronunciation was the main drive behind the
development of the phonetic alphabet. Perhaps you have already noticed that
some similar sounds have different representations in English orthography.
Take the following examples:

[f] as in trough, fun

[n] as in knife, night
[i] as in tiny, ceiling, tin

In addition, different sounds in many languages may be used in the writing

system with the same representation. Take the following English examples:

gh Pronounced as /f/ as in cough

Pronounced as /aʊ/ as in bough
ou Pronounced as /ɔ:/ as in bought
Pronounced as /ʌ/ as in tough

CHAPTER TWO 11
Language (all languages) changes over time. Spelling (orthography) is slower
to change than pronunciation. As a consequence, the discrepancy between
spelling and sounds gave impetus to a number of scholars to want to revise the
alphabet so that each sound would be represented by one and only one
symbol and each symbol would represent one and only one sound. Robert
Robinson (1617), Cave Beck (1657), Bishop John Wilkins (1668), Francis
Lodwick (1686) are some of the scholars who developed their own phonetic
writing systems. In 1888, the International Phonetic Association (IPA)
developed the most comprehensive phonetic chart which could be used to
symbolize the sounds that appear in all languages of the world. The main
characteristic of the phonetic alphabet is the one-to-one correspondence
between sound and symbol. In other words, each sound is represented by one
and only one symbol and each symbol represents one and only one sound.
Today, the phonetic alphabet is widely used to transcribe or write sounds in all
languages of the world. Take the following example:

EXAMPLE
English Orthography The man wanted to know the truth.
IPA Transcription /ðǝ mæn ˈwɒntɪd tǝ nǝʊ ðǝ tru:θ/

Each symbol in the ordinary orthography (alphabet) of a language is called a

letter. Each symbol in the phonetic alphabet is called a sound segment. As
such, the word truth is composed of five letters (t, r, u, t, h) but four phonemes
/t/, /r/, /u:/, /θ/. The ordinary writing in a language (i.e., use of letters) is called
orthography or writing while phonetic writing (i.e., use of phonemes) is usually
called transcription. The inventory of phonemes in the IPA phonetic alphabet is
so rich that it can be used to represent all sounds that appear in all languages
of the world. However, some but not all of these sounds appear in the English
language. The following table summarizes all the basic sounds that are
employed by English speakers in their speech (i.e., English vowels,
diphthongs, triphthongs, and consonants).

12 CHAPTER TWO
 SYMBOL EXAMPLE SYMBOL EXAMPLE
VOWELS /i:/ beat /ǝ/ an
/ɪ/ bit /u:/ boot
/e/ hen /ʊ/ book
/ɜ:/ bird /ɔ:/ four
/æ/ hat /ɑ:/ car
/ʌ/ cut /ɒ/ hot
DIPHTHONGS /ʊə/ poor /eɪ/ day
/ɔɪ/ boy /aʊ/ now
/ɪə/ here /aɪ/ dry
/əʊ/ go /eə/ hair
TRIPHTHONGS /aʊə/ power /ɔɪǝ/ loyal
/aɪə/ liar /ǝʊǝ/ mower
/eɪǝ/ layer
CONSONANTS /p/ pot /ʃ/ she
/b/ bat /ʒ/ vision
/t/ top /ʧ/ chat
/d/ desk /ʤ/ judge
/k/ cut /m/ man
/g/ good /n/ not
/f/ five /ŋ/ thing
/v/ van /r/ rat
/θ/ think /l/ look
/ð/ this /w/ one
/s/ sit /j/ yard
/z/ zip /h/ hat

Table 2.1: IPA symbols for British English vowels, diphthongs, triphthongs,
and consonants.

CHAPTER TWO 13
 SYMBOL EXAMPLE SYMBOL EXAMPLE
VOWELS /i/ beat /ǝ/ an
/ɪ/ bit /u/ boot
/e/ hen /ʊ/ book
/ɚ/ bird /ɔ/ four
/æ/ hat /ɑ/ car
/ʌ/ cut /a/ hot
DIPHTHONGS /ʊɚ/ poor /eɪ/ day
/ɔɪ/ boy /aʊ/ now
/ɪɚ/ here /aɪ/ dry
/əʊ/ go /eɚ/ hair
CONSONANTS /p/ pot /ž/ vision
/b/ bat /č/ chat
/t/ top /ǰ/ judge
/d/ desk /m/ man
/k/ cut /n/ not
/g/ good /ŋ/ thing
/f/ five /r/ rat
/v/ van /l/ look
/θ/ think /w/ one
/ð/ this /ʍ / why
/s/ sit /j/ yard
/z/ zip /h/ hat
/š/ she

Table 2. 2: IPA symbols for American English vowels, diphthongs, and

consonants.

In addition to phonemes, a number of other symbols are used in the phonetic

alphabet to represent the intricacies and nuances of sounds such as length,
aspiration, etc. These symbols are normally put either above (superscript) or

14 CHAPTER TWO
below (subscript) phonemes and indicate the features of phonemes. These
symbols are called diacritics or sporadic features. Arabic is a language which
uses sporadic features even in its ordinary orthography. In ordinary English
orthography, too, a limited number of diacritics are used (e.g., ℃).

Table 2.2 summarizes the most frequent diacritic symbols which are used in
IPA phonetic alphabet:

SYMBOL MEANING LOCATION

: long right of phoneme
◦ voiceless under phoneme
∙ syllabic under phoneme
h aspirated right superscript
w rounded right superscript
~ nasalized above phoneme
ˌ syllabic under phoneme
Table 2.3: Sporadic symbols used in phonetic transcription in IPA

In addition to these, phonologists often use cover symbols to refer to classes of

sounds. They use a capital C to refer to the class of consonants, a capital V to
represent vowels, a capital L for liquids, a capital N for nasals, and a capital G
for glides:

SYMBOL MEANING SYMBOL MEANING

C: long C Ṽ nasalized V
V: long V V́ stressed V
L̥ voiceless L Ņ syllabic N
G̥ voiceless G Ļ syllabic L
V̥ voiceless V

Phoneticians also repeat these symbols when they want to show a cluster of
phonemes. For example, to show a consonant cluster, the symbol C is
repeated.

CHAPTER TWO 15
 CC (cluster of 2 consonants) CCC (cluster of 3 consonants)

4. BROAD AND NARROW TRANSCRIPTION

An interesting point about phonemes is that they are not always pronounced
with the same pronunciation. In fact, the co-text and environment of the
occurrence of a phoneme plays a crucial role in the way the phoneme is
pronounced. All of us have the experience of changing the sound /n/,
unconsciously of course, into the sound /m/ in such Persian words as /ʃænbe/
(meaning Saturday) so that the word is pronounced as /ʃæmbe/. In fact, the /b/
sound following the /n/ sound causes this pronunciation difference. For the
most part, these pronunciation differences are surface phenomena. In other
words, our brains form the exact pronunciations of words (i.e., similar to those
found in standard dictionaries). When the brain orders the vocal organ to
vocalize these words, the physiological shortcomings of the human vocal organ
cause these pronunciation differences. Many phonologists use the phrase
'Ease of Pronunciation' to refer to this physiological phenomenon.

We can now conceptualize two types of pronunciations: phonetic and

phonemic. Phonemic representation refers to the pronunciation of words as
they exist in our minds; phonetic representation refers to the pronunciation of
words as they are actually pronounced by our tongues. We should, however,
be aware that only a very limited number of phonemes have different phonemic
and phonetic representations. Take the phoneme /p/ in English as an example.
When this phoneme appears in word-initial contexts, it is pronounced with a
puff of air. This phenomenon is known as aspiration. In non-word-initial
contexts, however, the phoneme /p/ is reduced to a phoneme which stands
between the phonemes /p/ and /b/.

Phonologists have developed two types of phonetic writing system to capture

these differences: (a) one in which only the mental (phonemic) representation
of phonemes is shown, and (b) one in which the actual-speech (phonetic)
representation of phonemes is shown. The former is called broad transcription

16 CHAPTER TWO
while the latter is called narrow transcription. Broad transcription only utilizes a
basic set of symbols. Narrow transcription utilizes the same set of symbols with
the addition of diacritics and other symbols. The second difference between
broad and narrow transcriptions is that phonemes represented in broad
transcription are put between two slant lines // whereas phonemes represented
in narrow transcription are put inside square brackets []. Take the following
examples:

SPELLING BROAD SYMBOL NARROW SYMBOLS

p /p/ [p] and [ph]
l /l/ [l] and [ɫ]

The different representations of a phoneme in narrow transcription are called

the allophones of that phoneme. Take the following examples:

(a) The phoneme /p/ has two allophones: unaspirated [p] and aspirated [ph].

EXAMPLE BROAD NARROW

spit /spɪt/ [spɪt]
Peter /ˈpi:tǝ/ [ˈphi:tǝ]
slap /slæp/ [slæp]

(b) The phoneme /l/ has two allophones: light [l] and dark [ɫ].
(c) The phoneme /k/ has two allophones: aspirated [kh] and unaspirated [k].

EXAMPLE BROAD NARROW

look /lʊk/ [lʊk]
black /blæk/ [blæk]
kill /kɪl/ [khɪɫ]

Allophones of a phoneme are in complimentary distribution. That is, they

cannot occur in the same context. For example, [ph] comes at the beginning of
a word while [p] occurs in other contexts.

CHAPTER TWO 17
Another point to notice about allophones is that the differences between them
are phonetic rather than phonemic. A phonetic difference does not cause a
change in meaning. A phonemic difference, however, brings about a change in
meaning. For instance, the difference in words like ship /ʃɪp/ and sheep /ʃi:p/ is
phonemic because these two words have two different meanings. Therefore,
we cannot consider /ɪ/ and /i:/ to be allophones of a basic phoneme. Words like
sheep /ʃi:p/ and ship /ʃɪp/ are called minimal pairs. Traditionally, minimal pairs
were defined as pairs of words that differ in one and only one phoneme. Take
the following examples:

thy /ðaɪ/ thigh /θaɪ/

ship /ʃɪp/ sheep /ʃi:p/
bow /baʊ/ wow /waʊ/
kill /kɪl/ keel /ki:l/

If one of the words in a minimal pair is repeated, a minimal set will result. Take
the following examples:

thy thigh thigh

ship ship sheep
bow wow bow
keel kill kill

One classic book with a good number of minimal pairs in it is the American PDs
(or American Pronunciation Drills). The American PDs is still widely used in
phonetics classes in a good number of language schools throughout the world.

It should be noted that some phonologists are inclined to use the diacritic
symbols that represent primary and secondary stress in ordinary writing too. In
this case, the symbol ˊ is used to represent primary stress and the symbol ˋ to
represent secondary stress. In phonetic transcription, however, the symbol ˈ is
used for primary stress and the symbol ˌ for secondary stress. The other
difference is that in ordinary writing, the symbols appear over the vowels that

18 CHAPTER TWO
carry them. In phonetic transcription, on the other hand, the ˈ symbol is put at
the top-left corner and the symbol ˌ at the bottom-left corner of the syllables that
carry them in British English. In American English, the ˈ symbol appears at the
top-right corner and the symbol ˌ at the bottom-right corner of the syllables that
carry them. Take the following examples:

WRITING BRITISH METHOD AMERICAN METHOD

dèmonstrátion /ˌdemǝnˈstreɪʃǝn/ /demˌǝnstreɪˈʃǝn/
àristócracy /ˌærɪˈstɒkrǝsɪ/ /ærˌɪstɒkˈrǝsɪ/
hèrmeneútics /ˌhɜ:mɪˈnju:tɪks/ /hɚˌmɪnu:ˈtɪks/
(The /ɚ/ sound in American English stands for the /ɜ:/ sound of British
English, and is pronounced as a combination of /ǝ/ and /r/ sounds).

5. SYLLABLE STRUCTURE

The sounds that result from one chest pulse form a syllable. In its minimal form,
a syllable consists of a vowel. In addition to the vowel a syllable may consist of
one or more consonants that appear on either or both sides of the vowel. In
some languages like Japanese, most often the syllable is composed of one
consonant followed by one vowel. These languages are called syllabic
languages. In syllabic languages, each syllable is represented by a symbol
(called syllabary) in the writing system. The word TOYOTA from the Japanese
language for example includes three syllables: TO, YO, and TA. Therefore, the
syllable structure of most Japanese syllables is very simple: Consonant +
Vowel (CV). Most languages are, however, alphabetic in that symbols (called
characters or letters) in their orthography represent sound segments or
phonemes rather than syllables. In these languages, the consonants and
vowels are arrayed in a linear fashion to represent the syllables, words,
sentences, etc. Arabic and Hebrew, however, tend to arrange their consonants
in a linear fashion, and superscribe or subscribe their vowels as diacritics or

CHAPTER TWO 19
sporadic features above or under their consonants. As such, Arabic and
Hebrew can ironically be called betagamic rather than alphabetic languages.

Many of the most famous languages of the world, including English, are,
however, alphabetic in the sense that they represent both their vowels and
consonants in the form of letters in their orthography. In such languages, words
are composed of one or more syllables. A syllable is a phonological unit
composed of one or more phonemes. Every syllable has a nucleus, which is
usually a vowel (but which may be a syllabic liquid or nasal). The nucleus may
be preceded by one or more phonemes called the syllable onset and followed
by one or more segments called the coda. English is an alphabetic language
which has a complex syllable structure. The syllable structure of English has
been presented in table 2.4 below.

SYLLABLE STRUCTURE EXAMPLE

V I /aɪ/
CV key /ki:/
CCV tree /tri:/
CCCV spree /spri:/
CVC seek /si:k/
CCVC speak /spi:k/
CCCVC scram /skræm/
CCCVCC striped /straɪpt/
VC an /æn/
VCC ant /ænt/
VCCC ants /ænts/
CVCC pant /pænt/
CVCCC pants /pænts/
CCCVCCC splints /splɪnts/
CCVCC stamp /stæmp/
Table 2.4: Syllable structure of English

20 CHAPTER TWO
Table 2.4 has used the symbol C to represent consonants and the symbol V to
represent vowels. Notice that the syllable structure of English includes at least
fifteen different types of syllables:

In this connection, it is interesting to notice that in alphabetic languages, the

number of vowels that appear in a word can be used as an index for
determining the number of syllables that make that word. A close look at the
syllable structures presented in table 2.4 above reveals that, in English,
consonant clusters can occur in both syllable-initial and syllable-final positions
(i.e., as onset or coda). Moreover, consonant clusters are not limited to two
consonants in English. In a word like street three consonants cluster together
at the beginning of the syllable to produce a CCCVC syllable. Another
interesting observation is that vowels can initiate syllables in English.

The syllable structure of Persian is, however, different. On the one hand,
Persian syllables cannot be initiated with vowels; even words that seem to start
with a vowel include the glottal stop /ʔ/ as the syllable onset. On the other
hand, syllable-initial consonant clusters are impossible in Persian. In addition,
syllable-final consonant clusters in Persian normally take no more than two
consonants in their structure. As such, most Persian syllables belong in one of
the three syllable structures (i.e., CV, CVC, or CVCC) presented in table 2.5
below. Take the following Persian examples:

PERSIAN SYLLABLES EXAMPLE

CV ba /ba:/ meaning with
CVC toop /tu:p/ meaning ball
CVCC satr /sætr/ meaning line
abr /ʔæbr/ meaning cloud
Table 2.5: Persian syllable structure

The differences between the syllable structure of Persian and English are
responsible for a good portion of Iranian EFL learners' pronunciation problems.

CHAPTER TWO 21
In fact, many Iranian EFL learners tend to insert the vowel /e/ in many mono-
syllabic English words to make them readily pronounceable. In addition, since
Persian syllables cannot be initiated by vowels, many Iranian EFL learners start
pronouncing vowel-initial English syllables with the consonant /ʔ/. The term
Penglish is sometimes used to refer to Persian pronunciation of English words.
The result of such mispronunciations is that many monosyllabic English words
are rendered as bi- or tri-syllabic by some Iranian EFL students. Take the
following examples:

EXAMPLE PRONUNCIATION MISPRONUNCIATION

out /aʊt/ /ʔaʊt/
tree /tri:/ /teri:/
dress /dres/ /deres/
street /stri:t/ /ʔesteri:t/
sky /skaɪ/ /ʔeskaɪ/

6. RECEIVED PRONUNCIATION

Why should anyone want to learn the speech sounds for a British accent that is
spoken by less than 3% of the population of that country? And, Britain itself
provides only a minority of the English speaking peoples of this world. The
reason is mainly to do with a legacy of history. Throughout the nineteenth
century and throughout the early part of the twentieth century, Received
Pronunciation (RP) was very much the language of the ruling and educated
classes. A vicious circle was then instituted: those who could afford an
education went to the private schools and to university where they learnt RP;
the teachers for the next generation were then drawn from this class of people
to teach the next generation of the ruling elite. Thus, the educators were
instructed in RP to teach RP. Therefore, most of the early phonetics work was
carried out by RP speakers using their own accent as the "standard" from
which all other varieties were measured.

22 CHAPTER TWO
RP does have the advantage of being a regionless accent although many of
these speakers are concentrated in the southeast of England. This is partly to
do with its circulation amongst the educated and elite, but it was also the
variety that was exported through the colonies during the time of the British
Empire. Consequently, the ruling elites of many of these countries also adopted
RP as their "standard."

One of the problems of RP is that its association with the ruling elite has meant
that it is regarded as being a classist accent. Aloofness and snobbishness are
characteristics often associated with RP speakers. The legacy of history has
meant that RP is the closest that British English has towards a standard
variety. It has been subjected to a great deal of academic scrutiny, it is the
choice of many broadcasters and it is an accent that many parents aspire for
their children to emulate. The main advantage of learning and using RP is that
it is an accent that all English speakers can understand, wherever they are in
the world. It may have many problems attached to it, but it is still the most
widely understood and respected of the British English varieties. The American
counterpart of RP is usually referred to as General American English (GAE).
Foreign language learners should make no value judgments on the advantage
of learning one (RP or GAE) over the other.

EXERCISE

1. Write the phonetic symbol for the first sound in each of the following words
according to the way you pronounce it.

Examples: ooze [u:] psycho [s]

write [ ] advantage [ ] counter [ ]

mnemonic [ ] variety [ ] accent [ ]

philosophy [ ] standard [ ] post [ ]

read [ ] still [ ] cheat [ ]

CHAPTER TWO 23
2. Write the following words in both broad and narrow phonetic transcriptions,
according to your pronunciation.

WORD BROAD TRANSCRIPTION NARROW TRANSCRIPTION

rumor

pacific

presumptuous

characteristic

redemption

soliloquy

catastrophe

posthumous

realization

rupture

3. The following lines of poetry include mispronunciations. Rewrite them in the

correct narrow phonetic transcription.

MISPRONUNCIATIONS CORRECT NARROW TRANSCRIPTIONS

[ðə thaɪm hæz cʌm]

[ði: vɔlrəs seɪd]

[thu: tɔ:lk əv mænɪ θɪŋz]

[əv ʃu:z ænd ʃɪpz]

[ænd si:lɪŋ wæx]

24 CHAPTER TWO
4. The following sentences are written in normal English orthography. Write
them in broad transcription.

ORTHOGRAPHY BROAD TRANSCRIPTIONS

How was your weekend?

Have you ever seen him?

He does some incredible things.

I've never been there.

How many rooms are there?

How do you like it there?

You're in great shape, Keith.

I won't play too hard.

I hardly ever do yoga at noon.

How often do you exercise?

5. Write the onset of each of the following syllables in the spaces provided. If
the syllable has no onset, write Ø. The first one has been done as an
example for you.

SYLLABLE ONSET SYLLABLE ONSET

e.g.: street str pant

speak ant

scram striped

an splints

ants pants

CHAPTER TWO 25
6. Write the coda of each of the following syllables in the spaces provided. If
the syllable has no coda, write Ø. The first one has been done as an
example for you.

SYLLABLE CODA SYLLABLE CODA

e.g.: street t pant

speak ant

scram striped

an splints

ants pants

7. Define the following terms briefly.

complimentary distribution allophone

ease of Pronunciation broad transcription
diacritics narrow transcription
sporadic features coda
transcription onset
sound segment syllabary
phonetic alphabet syllable
phoneme minimal pair

26 CHAPTER TWO
 CHAPTER THREE
TRADITIONAL ARTICULATORY PHONETICS

1. INTRODUCTION

Early phoneticians used phonetic symbols as abbreviations for full descriptions

of the sounds of language. In order to afford a precise account of the sounds of
language, they used the vocal tract activities during sound articulation as their
classification criterion. In other words, each sound in any language is
considered to be a complex pattern of overlapping waves moving through air.
These patterns are thought to be caused by different configurations of the
various organs of speech as air travels from the lungs out through the nose or
mouth or both.

So, phoneticians argued that a sound can be best described if we describe it in

terms of air modifications that occur in the vocal organ of speakers. In fact,
some phoneticians turned to anatomists for help in this connection. This gained
so much importance that a good number of books on phonetics included
chapters devoted partially or completely to a discussion of the functions of
human respiratory system. A discussion of the human vocal organ is, therefore,
necessary here.

The primary function of the human vocal organ is not speech production. In
fact, this organ has been developed for purposes of breathing. In other words,
the primary function of this organ is respiration. We take oxygen into our lungs
(inhalation) and send carbon dioxide out of our lungs (exhalation) in order to
remain alive. Speech is only the peripheral or subsidiary function of human
vocal organ also called the human vocal tract. Speech sounds are not the only
sounds produced by human vocal tract. All of us sneeze, snore, grunt, groan,

CHAPTER THREE 27
scream, hiccup, cough, etc. It is, however, impossible to find any language in
which these noises are part of its speech articulation system.

Sagittal section of human vocal tract (Reprinted from Lieberman and Blumstein, 1988,
P. 43)

In order to produce speech, we modify the flow of air that passes between our
lungs and the outside atmosphere. These modifications are normally applied to
the flow of carbon dioxide during exhalation to produce a class of sounds
technically called egressive sounds. In some African languages, however,
people make some speech sounds when they are inhaling air (ingressive

28 CHAPTER THREE
sounds). It can, therefore, be concluded that air stream mechanism is at the
heart of speech.

Normal name Fancy name Adjective

lips labia labial
teeth ---- dental
alveolar ridge ---- alveolar
(hard) palate ---- palatal
soft palate velum velar
uvula ---- uvular
upper throat pharynx pharyngeal
voicebox larynx laryngeal
tongue tip apex apical
tongue blade lamina laminal
tongue body dorsum (back) dorsal
tongue root ---- radical
glottis ---- glottal
nose nasal cavity nasal
mouth oral cavity oral

In addition to their normal names, many of the parts of the vocal tract have
Latin or Greek fancy names. The adjectives we use to describe sounds made
with each part are usually based on the Latin/Greek name.

In phonetics, the terms velum, pharynx, larynx, and dorsum are used as often,
or more often, than the simpler names. Many of the names that appear in the
sagittal section of the human vocal tract are already known to you. There are,
however, some names that you may find new. It is, therefore, necessary to
provide a definition for each of these new terms.

CHAPTER THREE 29
Cartilages and intrinsic muscles of the larynx (Reprinted from Lieberman and Blumstein,
1988, P. 98)

The alveolar ridge is a short distance behind the upper teeth. It is a change in
the angle of the roof of the mouth. (In some people it's quite abrupt, in others
very slight.) Sounds which involve the area between the upper teeth and this
ridge are called alveolars. Hard palate is the hard portion of the roof of the
mouth. The term "palate" by itself usually refers to the hard palate. Soft
palate/velum is the soft portion of the roof of the mouth, lying behind the hard
palate. The velum can also move: if it lowers, it creates an opening that allows
air to flow out through the nose; if it stays raised, the opening is blocked, and
no air can flow through the nose. Uvula is the small, dangly thing at the back of
the soft palate. The uvula vibrates during the /r/ sound in many French dialects.
Pharynx is the cavity between the root of the tongue and the walls of the upper
throat. Tongue blade is the flat surface of the tongue just behind the tip.
Tongue body/dorsum refers to the main part of the tongue, lying below the hard
and soft palate. The body, specifically the back part of the body (hence

30 CHAPTER THREE
"dorsum", Latin for "back"), moves to make vowels and many consonants.
Tongue root is the lowest part of the tongue in the throat. Epiglottis refers to the
fold of tissue below the root of the tongue. The epiglottis helps cover the larynx
during swallowing, making sure that food goes into the stomach (through the
esophagus) and not into the lungs. A few languages use the epiglottis in
making sounds. English is fortunately not one of them. Vocal folds/vocal cords
are folds of tissue (i.e., membranes) stretched across the airway to the lungs.
They can vibrate against each other, providing much of the sound during
speech. Glottis refers to the opening between the vocal cords. During a glottal
stop, the vocal cords are held together and there is no opening between them.
Larynx is the structure that holds and manipulates the vocal cords. The
"Adam's apple" in males is the bump formed by the front part of the larynx.

As it was mentioned in the previous chapter, human beings can not only
produce sounds but also noise by means of their vocal tracts. All of us have the
experience of coughing after catching cold or flu. We also sneeze, snore, grunt,
groan, scream, etc. You, however, cannot find any one language in which
words are formed as sequences of these noises. There is a quite simple
reason for this. Noises interfere with the primary functions of our vocal organs,
that of respiration. In addition, they cause pain and bruise in different parts of
the vocal tract. Now, you can imagine what happens to the health of an
individual if he had to use these noises in his speech. In fact all natural
languages employ sounds (i.e., consonants and vowels). These are well-
organized sounds which do not cause problems for respiration.

In order to produce speech sounds, the flow of pulmonary air in the vocal tract
must somehow be modified. These modifications are of four main types: (1)
complete blockage followed by sudden release, (2) complete blockage followed
by gradual release, (3) constriction or narrowing, and (4) no blockage at all. In
fact, these modification types have been used in traditional phonetics as the
criteria for the classification of speech sounds. These modifications are
technically referred to as manner of articulation. These modifications do not

CHAPTER THREE 31
occur at one single part of the vocal tract. Rather, they may happen at any part
of the super-glottal (from the glottis upwards) section of the vocal tract. These
places (sometimes called articulators) include the lips, the teeth, the alveolar
ridge, the palate, the velum, and the glottis. Place of articulation is the technical
term which is used to refer to the places along the vocal tract at which air
modifications take place. In addition to place and manner of articulation,
speech sounds can be voiced or unvoiced (voiceless) depending on whether
the vocal cords are set into vibration by the impact of pulmonary air. This
phenomenon is technically referred to as voicing.

2. SPEECH ARTICULATION

Voicing, manners, and places of articulation are responsible for the production
(articulation) of different speech sounds. Sound is produced by the interference
of the flow of air through the mouth (and nose). Consonants are created when
the airflow is directly restricted, or obstructed. As such, pulmonary air cannot
escape from the oral cavity without creating audible friction. By way of contrast,
vowels are created when the airflow is not crucially restricted or obstructed.
Therefore, pulmonary air can escape the oral cavity without creating audible
friction. Defining characteristics of consonants include: (a) voicing (b) place of
articulation (c) manner of articulation. Defining characteristics of vowels are: (a)
lip rounding, and (b) relative tongue position.

2.1. MANNER OF ARTICULATION

Manner of articulation refers to the nature of the obstruction of pulmonary air

flow. In order to fully appreciate the differences among speech sounds, as well
as indicating the place of articulation, it is necessary to determine the nature
and extent of the obstruction of airflow involved in their articulation. The type of
airflow obstruction is known as the manner of articulation. The manner of
articulation is particularly defined by four major factors: (a) whether there is
vibration of the vocal cords (voiced vs. voiceless), (b) whether there is

32 CHAPTER THREE
obstruction of the airstream at any point above the glottis (consonant vs.
vowel), (c) whether the airstream passes through the nasal cavity in addition to
the oral cavity (nasal vs. oral), and (d) whether the airstream passes through
the middle of the oral cavity or along the side(s) (non-lateral vs. lateral). An
example of this can be found by looking at the following words:

nine /naɪn/ dine /daɪn/ line /laɪn/

They all begin with voiced, alveolar consonants /n/, /d/, and /l/. Yet, they are all
clearly different in both sound and meaning. The kinds of constriction made by
the articulators are what make up this further dimension of classification. There
are two common kinds of constriction that often occur in English: plosive and
fricative. Also, there are other less common constrictions: nasal, affricate,
lateral, and approximant. Traditional phonetics, however, used three cover
terms to refer to all kinds of constriction: plosive, fricative, and affricate. In the
following sections, these manners of articulation will be discussed with greater
detail so that the reader can fully understand what they mean.

2.1.1. OCCLUSIVE AND PLOSIVE

Occlusives require a complete closure of the speech canal, not just a

restriction. This distinguishes them from the continuants. The occlusion is
twofold: (a) the airstream is halted by a sudden closure in the oral cavity; (b)
the trapped air is freed by abruptly releasing the closure. If the trapped air is
gradually released, an affricate consonant is articulated. Occlusives in English
include /p/, /b/, /m/, /t/, /d/, /n/, /k/, /g/, and /ŋ/.

[p] is a voiceless bilabial stop consonant. The lips are pressed tightly together.
The air is trapped behind the lips. The vocal cords are kept far apart, and the
nasal cavity is closed by the velum. Then the trapped air is suddenly released.
[b] is the voiced counterpart of [p]. The only difference is that the vocal cords

CHAPTER THREE 33
are close to each other and vibrate during the articulation of [b]. In the case of
/m/, the nasal cavity is open.

/b/ and /p/ /t/ and /d/ /k/ and /g/

[t] is a voiceless dental or alveolar stop. The tongue makes contact with the
front teeth or with the alveolar ridge directly above them. There is no vocal cord
vibration and the nasal cavity is blocked. [d] is a voiced dental or alveolar stop.
It is produced in the same way as [t] but with vibration of the vocal cords. In the
case of /n/, the nasal cavity is open to let the air pass through it.

[k] is a voiceless velar stop. With the tongue tip resting against the lower teeth,
the back of the tongue makes contact with the soft palate. [g] is its voiced
counterpart. Its articulation is the same as [k], but with vibration of the vocal
cords. The corresponding velar nasal [ŋ] is usually voiced as well.

Some languages, including Persian, have a glottal occlusive [ʔ] too. The glottal
stop can be produced in either of the two ways: (a) by the sudden opening of
the glottis under pressure from the air below, or (b) by the abrupt closure of the
glottis to block the airstream. The glottal stop is always voiceless, as the
complete closure of the vocal cords precludes their vibration.

Occlusives can be categorized into two major types: stops and plosives. The
two categories are in inclusional distribution. That is, all plosives are stops but
all stops are not necessarily plosive. This relationship can be schematically
represented as:

34 CHAPTER THREE
 (PLOSIVE ➙ STOP).

Plosive sounds are made by forming a complete obstruction to the flow of air
through the mouth and nose. The first stage is that a closure occurs. Then the
flow of air builds up and finally the closure is released, making an explosion of
air that causes a sharp noise. Try to slowly say /p/ to yourself. You should be
able to feel the build up of air that bursts into the /p/ sound when you open your
lips. It should be noted that a plosive cannot be prolonged or maintained so
that once the air has been released, the sound has escaped. As such, plosive
sounds lack the length feature. Contrast this quality of plosives with a fricative
in which you can lengthen the sound. The plosive sounds in RP are: /b/, /p/, /t/,
/d/, /k/, and /g/. As it was mentioned earlier, plosive sounds belong to a more
general class of sounds called stops. A stop sound is one in which the flow of
air is completely blocked only in the oral cavity. Stops also include such sounds
as /m/, /n/, and /ŋ/. Take the following examples:

moon /mu:n/ night /naɪt/ thing /θɪŋ/

You can feel that in the production of such sounds as /m/, /n/, and /ŋ/ the flow
of air is completely blocked in the mouth. However, air can flow through the
nose. As such, the air cannot burst into these sounds because they can be
lengthened. In addition to these sounds, /ʧ/ and /ʤ/ are also marked by a
complete blockage of pulmonary air in the oral cavity. Here, again, the
blockage is not followed by an abrupt release. Rather, the blocked air is
gradually released to create friction. Some phoneticians rank these two sounds
among the stop consonants while many others classify them as affricates. Take
the following examples:

Jack /ʤæk/ chat /ʧæt/

2.1.2. FRICATIVE

A fricative is the type of consonant that is formed by forcing air through a

narrow gap in the oral cavity so that a hissing sound is created. Typically air is

CHAPTER THREE 35
forced between the tongue and the place of articulation for the particular sound.
Try it yourself. Say the /f/ in fin /fɪn/, the /θ/ in thin /θɪn/ and the /ʃ/ in shin /ʃɪn/.
You should be able to feel the turbulence created by the sounds. It is possible
to maintain a fricative sound for as long as your breath holds out. This is very
different from a plosive sound. Other fricatives include the /v/ in van, the /s/ in
sin, the /h/ in hat /hæt/, the /ð/ in that /ðæt/, the /z/ in zoo /zu:/ and the /Ʒ/
sound in genre /ˈʒɑ:nrə/.

Fricative consonants result from a narrowing of the speech canal that does not
achieve the full closure characteristic of the occlusives. The shape and position
of the lips and/or tongue determine the type of fricative produced. Phoneticians
usually distinguish between so-called true fricatives and the related class of
spirants. During the production of a fricative, the airstream can be directed in
several ways. First, in the case of true fricatives, the tongue channels the air
through the center of the mouth (like in the case of the dorsal fricatives).
Second, the tongue can also channel the air down the side(s) of the mouth (like
in the case of the lateral fricatives). Finally, in the case of labial and dental
fricatives, the shape and position of the tongue is not important. This makes
sense because the place of articulation is not, strictly speaking, in the oral
cavity at all.

/f/ and /v/

36 CHAPTER THREE
[f] is a voiceless labiodental fricative. The lower lip is brought close to the upper
teeth, occasionally even grazing the teeth with its outer surface, or with its inner
surface, imparting in this case a slight hushing sound. Considering its place of
articulation, it is unimportant to class this sound as dorsal or lateral fricative. [v]
is a voiced labiodental fricative. Its articulation is the same as [f], but with
vibration of the vocal cords. Like [f], considering its place of articulation, it is
unimportant to class [v] as dorsal or lateral fricative.
Among the fricatives are ones described as hissers and hushers. The
realization of a hisser requires a high degree of tension in the tongue: a groove
is formed along the whole length of the tongue, in particular at the place of
articulation where the air passes through a little round opening. English hissers
are /s/, /z/, /ʧ/, and /ʤ/. The hushers are produced similarly, but with a
shallower groove in the tongue, and a little opening more oval than round. The
lips are often rounded or projected outwards during the realization of a husher.
English hushers are /ʃ/, and /ʒ/.

HISSER HUSHER

Spirants involve the same restriction of the speech canal as fricatives, but the
speech organs are substantially less tense during the articulation of a spirant.
Rather than friction, a resonant sound is produced at the place of articulation.

[s] is a voiceless alveolar fricative (also a hisser). This apico-alveolar hisser is

produced by bringing the end of the tongue close to the alveolar ridge. Hissers
like /s/ can be divided into three categories, according to the precise part of the
tongue that comes into play: (a) coronal hissers which involve the front margin

CHAPTER THREE 37
of the tongue (as in English), (b) apical hissers which involve the very tip or
apex of the tongue (as in Castilian Spanish), and (c) post-dental hissers where
the front part of the tongue body is involved (as in French). The quality of the
sound is noticeably altered in these three types of hissers. The IPA uses
diacritical marks to indicate distinctions of this magnitude. [z] is a voiced
alveolar fricative (and a hisser). The same mechanism that produces [s] also
produces [z] but with vibration of the vocal cords. In general, the remarks made
for the voiceless sound are equally valid for the voiced variant. Other hissers
are [ʧ] and [ʤ]. [ʧ] is a voiceless palatal fricative. The tongue body forms a
groove and approaches the hard palate. In terms of general tongue shape, this
articulation qualifies as a hisser. [ʤ] is a voiced palatal fricative. It is articulated
in the same way as [ʧ] but with vibration of the vocal cords. [ʃ] is a voiceless
alveolar fricative (and a husher). The tip of the tongue touches the alveolar
ridge. The groove in the center of the tongue is shallower than is the case for
hissers. [ʒ] is its voiced counterpart.

Basically, friction and fricatives develop from tense articulations. When the
articulation is lax, resonance, and thus a spirant, occurs. Also realize that many
spirants can be thought of as the lax counterparts of stop consonants. English
interdentals /ð/ and /θ/ are spirants.

/θ/ and /ð/

38 CHAPTER THREE
[θ] is a voiceless dental spirant. The tongue tip is held close to the upper teeth,
either behind them (dental) or just underneath them (the interdental
articulation). This spirant is the lax counterpart of the stop [t]. Considering its
place of articulation, it is unimportant to class this sound as a dorsal or lateral
fricative. [ð] is the voiced counterpart of /θ/. Its articulation is the same as /θ/
but with vibration of the vocal cords. This spirant is the lax counterpart of the
stop [d].

FRICATIVE SPIRANT LATERAL

Laterals are generally considered to be a special case, since physically

speaking they could be grouped among the fricatives and spirants. They are
called laterals since, during their production, the back of the tongue makes
contact with the hard palate while the front of the tongue sinks down,
channeling the air laterally around the tongue, down the side (or sometimes
both sides) of the mouth. On the other hand, for non-lateral articulations, the
back of the tongue rests against the top molars, and the air flows over the
tongue down the center of the mouth. There are two distinct types of lateral: (a)
lateral fricatives, where the articulation, requiring a great deal of muscular
tension, resembles that of the fricatives (except for the position of the tongue),
and (b) non-fricative lateral, often called liquids, whose articulation is very close
to the spirants. It is interesting to note that the location of the lateral channel
through which the air flows is unimportant. That is, whether it is on the left, the
right, or both sides of the mouth, the nature of the sound produced is
unchanged. English laterals are usually non-fricative. In addition, no distinction

CHAPTER THREE 39
is made between voiceless and voiced variants because it is very rare for a
language to distinguish laterals according to voice. English liquids are /r/ and
/l/.

2.1.3. LATERAL

To produce a lateral sound, air is obstructed by the tongue at a point along the
centre of the mouth but the sides of the tongue are left low so that air can
escape over its sides. In fact, the tongue is strongly flexed and the air is forced
through a narrow oval cavity, producing a hushing sound. /l/ is the clearest
example of a lateral sound in English. Both the clear [l] and the word-final dark
[ɫ] allophones (i.e., the variants of the same phoneme) of /l/ are lateral sounds.
When an alveolar plosive is followed by the lateral /l/, then what happens is that
we simply lower the sides of the tongue to release the compressed air, rather
than raising and lowering the blade of the tongue. If you say the word 'bottle' to
yourself you can feel the sides of the tongue lower to let out the air.

[l] [ɬ]

2.1.4. APPROXIMANT

An approximant is a consonant that makes very little obstruction to the airflow.

Approximants are divided into two main groups: semivowels (also called glides)
and liquids. The semivowels are /h/ as in hat /hæt/, /j/ as in yellow /ˈjeləʊ/, and

40 CHAPTER THREE
/w/ as in one /wʌn/. They are very similar to the vowels /ɜ:/, /u:/ and /i:/,
respectively. However, semivowels are produced as a rapid glide. The liquids
include the lateral /l/ and /r/ sounds in that these sounds have an identifiable
constriction of the airflow, but not one sufficiently obstructive enough to
produce a fricative sound. Approximants are never fricative and never
completely block the flow of air.

closed glottis posterior shunt (in /h/) open glottis

Adjustments of the larynx for phonation (Reprinted—with some modifications—from

Lieberman and Blumstein, 1988, P. 111)

[h] is an interesting case. It is a voiceless glottal fricative. The glottis is almost

completely closed, except for a narrow opening (i.e., the posterior shunt) in its
upper part at the level of the arythenoidal cartilage. A strong friction develops
when air flows through this opening.

One point of caution is, however, necessary here. English and Persian make
use of different forms of the approximant /r/. On the whole, the approximant /r/
is of three types: (a) flap, (b) trill, and (c) retroflex. English draws on trill and
retroflex types while Persian employs trill and flap types. In Persian, the /r/
sound becomes flap whenever it appears between two vowels. A flap /r/ is
articulated when the blade of the tongue touches the roof of the mouth only
once. Try saying the following Persian words to yourselves:

/bɑ:ˈrɑ:n/ (meaning rain) /ˈmibɑ:ræd/ (meaning precipitates)

CHAPTER THREE 41
You should be able to feel the /r/ sound in these Persian words. It is somewhat
similar to the pronunciation of the intervocalic (between two vowels) /t/ sound in
American pronunciation of such words as 'water', 'butter', 'better', etc. Notice,
however, that English lacks flap /r/. The trill /r/ is articulated when the blade of
the tongue repeatedly touches the roof of the mouth for several times (similar
to the noise made by a chain gun). The /r/ sound in the Russian language is an
example of the extreme case of trill /r/. Both English and Persian employ trill /r/.
The retroflex /r/ is specific to English. Whenever the /r/ sound starts a word in
English, it becomes retroflex: (a) the tip of the tongue curls back, (b) the upper
and lower teeth approach each other, and (c) the lips are rounded and sent
forward. Hence, the retroflex /r/. try saying the English words 'write', 'red', etc.
to yourselves. They start with retroflex /r/.

RETROFLEX TAP TRILL

Many phoneticians use the term vibrant to refer to these and similar
consonants. Vibrant consonants involve one or more tapping or flapping
vibrations of the speech organs under pressure from the airstream. Part of the
tongue makes contact with the palate, most commonly at the alveolar ridge, the
soft palate, or (in some languages) the uvula. One or more very brief
occlusions (or air blockage) occur successively, accompanied by short
resonances. Vibrants are generally voiced. In narrow transcription, a small
subscript circle may be added to any IPA symbol to indicate a voiceless
variant. There are two distinct classes of vibrants: (a) those with only one

42 CHAPTER THREE
vibration, called taps, and (b) those with multiple vibrations, called trills. [ɾ] is an
alveolar tap (sometimes called flap). The alveolar region serves as the target
for the tongue tip, which vibrates there briefly before falling back to rest against
the lower teeth. [ɽ] is a retroflex tap. The tip of the tongue curves up and back,
and its underside vibrates briefly against the roof of the mouth, before falling
back to rest against the lower teeth. [r] is an alveolar trill. The alveolar region
serves as the target for the tongue tip, which vibrates there under pressure
from the airstream behind. The vibration produces occlusive sounds and
vocalic-type resonances in rapid alternation. This is the famous rolled /r/ of
Spanish, Russian, and some other languages.

2.1.5. NASAL

You have already noted in the discussion of plosives that, in the articulation of
some sounds, pulmonary air flows throw the nasal cavity to produce a nasal
consonant. A nasal consonant is a consonant in which air escapes only
through the nose. For this to happen, the soft dorsal part of the soft palate is
lowered to allow air to pass it, whilst a closure is made somewhere in the oral
cavity to stop air escaping through the mouth. You can feel if a sound is a nasal
sound or not by placing your hand in front of your mouth and feeling if any air is
escaping or not. There are three nasal sounds in English. The /m/ in mat /mæt/,
the/n/ in not /nɒt/ and the /ŋ/ in sing /sɪŋ/ or think /θɪŋk/. The velar /ŋ/ does not
occur in Persian. The Iranian learner of English, therefore, replaces the /ŋ/
phoneme by a sequence of two phonemes /n/, and /g/. In English, whenever
the letter n appears before the letters g and k, it is pronounced as /ŋ/. Nasal
sounds are considered to belong to the stop category along with plosives and
affricates.

The nasal “occlusives” of the vast majority of the world's languages are voiced.
Very few not-so-famous languages have voiceless nasals too. During the
production of these nasal “occlusives”, the soft palate is lowered to a greater or
lesser extent, allowing a portion of the airstream to pass through the nasal

CHAPTER THREE 43
cavity. Occlusion occurs in the mouth only; the nasal resonance is continuous.
Indeed, many linguists rank the nasals among the continuants. /m/ is a bilabial
nasal. The mouth is configured just as for the corresponding bilabial stop /p/
and /b/. The lips are pressed tightly together. The air builds up and is suddenly
released. /n/ is a dental or alveolar nasal. The mouth is configured just as for
the corresponding dental or alveolar stop /t/ and /d/. The tongue makes contact
either with the front teeth, or with the alveolar ridge directly above them. /ŋ/ is a
velar nasal. The configuration of the mouth is very close to that of the
corresponding velar stop /k/ and /g/. With the tongue tip resting against the
lower teeth, the back of the tongue makes contact with the soft palate. But as
the soft palate is lowered (to allow air to flow through the nasal cavity), the
tongue's movement is more important for the nasal than for the oral sound.

/m/ /n/ /ŋ/

2.1.6. AFFRICATE

An affricate is a plosive immediately followed by a fricative in the same place of

articulation.

The /ʧ/ in chap /ʧæp/ and the /ʤ/ in jeep /ʤi:p/ are the two clear affricates in
English. If you think about it, the /ʧ/ sound is made up from the plosive /t/ and

44 CHAPTER THREE
the fricative /ʃ/ sounds. Likewise, the /ʤ/ sound is made up from the plosive /d/
immediately followed by the fricative /z/.

/ʧ/ and /ʤ/

2.1.7. VOICING (LEVEL OF VIBRATION)

All of us inhale oxygen and exhale carbon dioxide. Pulmonary air gets out the
lungs and enters the bronchi. The two bronchi meet each other and form the
trachea. The trachea is intercepted by the voicebox or larynx on its way out.
Inside the voice box, there are two membranes that are hinged together at the
back. These membranes are called the vocal cords or the vocal folds. The
vocal cords can make a wedge-shaped opening when they are far apart. This
opening is called the glottis.

The level of vibration of the vocal cords determines whether a sound is voiced
or unvoiced. When the glottis is open, pulmonary air passes through it easily
without causing any friction. That is, if the vocal cords are apart, then air can
escape unimpeded. Sounds produced in this way are said to be voiceless. The
easiest example of this is to whisper. When you whisper, your glottis is wide
open and, therefore, all the sounds produced are voiceless. However, when the
glottis is closed, the vocal cords are set into vibration by the impact of the

CHAPTER THREE 45
pulmonary air. When the vocal cords vibrate, voice sounds are produced.
When they do not vibrate, voiceless sounds result. That is, if the vocal cords
are very close together, the air will blow them apart as it forces its way through.
This makes the cords vibrate, producing a voiced sound. Vocal cord vibration is
technically referred to as voicing. In English, only a limited number of
consonants are voiceless. /ʧ/, /s/, /p/, /k/, /f/, /ʃ/, /t/, /θ/, and /h/ are voiceless. All
other phonemes are voiced.

CLOSED GLOTTIS OPEN GLOTTIS

Adjustments of the larynx for phonation (Reprinted—with some modifications—from

Lieberman and Blumstein, 1988, P. 111)

To feel the distinction between voiced and voiceless sounds is very easy. Place
your finger and thumb lightly on your throat. Say ssssssss to yourself. Then say
zzzzzzz. Repeat these a few times. Then substitute fffffff and vvvvvvv sounds.
You should be able to feel the vibration of the cords when you say zzzzzz and
vvvvvv, but not when you say sssssss and fffffff. It is also possible to hear the
vibration through bone conduction. Instead of putting your fingers on your
throat, put your index fingers in your ears and repeat the above sounds. You
should hear a low buzzing sound when you articulate zzzzzz and vvvvvv, but
hear almost nothing for the other two sounds.

46 CHAPTER THREE
Voicing is important in a language like English because the meaning of a sound
often depends on whether that sound is voiced or not. For example, 'big'
carries a very different meaning from 'pig'. English has many sounds that are
paired up in this way. In fact, there are a number of cases in which the place
and manner of articulation of phonemes are the same, but the meaning is
dependant upon whether the sound is voiced or not.

2.1.8. FORTIS AND LENIS

Fortis and Lenis are closely related to level of vibration. Fortis sounds are those
that are made with strong muscular effort, originating in the lungs. Lenis is the
opposite. To fully understand what is meant by fortis and lenis, your attention is
drawn to the fact that all of us have the ability to whisper. In fact, during a
whisper, all voiced sounds are devoiced and yet still remain distinct from
voiceless sounds. Try whispering /pig/ and then /big/. You can still differentiate
the two because the intensity of the burst of air is greater for the fortis /p/ than
for the lenis /b/. Other fortis sounds include /t/ and /k/. Other Lenis sounds
include /d/ and /g/. In English, the fortis/lenis distinction happens to duplicate
the voiced/voiceless one found in minimal pairs. So all the voiced sounds are
lenis and all the voiceless sounds are fortis sounds.

2.1.9. LIP ROPUNDING

Another important form of air modification in which the lips are involved is lip
rounding. In the production of some vowels, the glide /w/, and the retroflex /r/,
the lips are rounded so that their shape becomes circular. This phenomenon is
called lip-rounding. It is a useful defining feature in the classification of vowels.

2.2. PLACE OF ARTICULATION

As it was noted earlier, the distinction between manner of articulation and place
of articulation is particularly important for the classification of consonants. The
place of articulation is the point where the airstream is obstructed. In general,
the place of articulation is simply that point on the palate where the tongue is

CHAPTER THREE 47
placed to block the stream of air. However, the palate is not the only place of
articulation.

The place of articulation can be any of these points: (a) the lips (labials and
bilabials), (b) the teeth (dentals), (c) the lips and teeth (labio-dentals—here the
tongue is not directly involved), (d) the alveolar ridge (that part of the gums
behind the upper front teeth—alveolar articulations), (e) the hard palate (given
its large size, one can distinguish between palato-alveolars, palatals and
palato-velars), (f) the soft palate (or velum—velar articulations), (g) the uvula
(uvulars), (h) the pharynx (pharyngeals), and (i) the glottis (glottals).

After the air has left the larynx, it passes into the vocal tract. Consonants are
produced by obstructing the air flow through the vocal tract. There are a
number of places where these obstructions can take place. These places are
known as the articulators. They include the lips, the teeth, the alveolar ridge,
the hard palate, the soft palate, and the throat. Some phoneticians define
articulators as the movable parts of the vocal tract.

2.2.1. LIPS

Some sounds are produced when the pulmonary air is modified (blocked or
constricted) by the lips. If both of the lips are used to articulate a sound, then it
is said to be a bilabial sound. Examples of bilabial sounds include /p/, /b/ and
/m/. Sometimes, the upper teeth and the lower lip collaborate to produce a
sound. In this case, the sound is said to be a labiodental sound. In English /f/
and /v/ are labiodental consonants.

2.2.2. TEETH

The two 'th' sounds of English (ð and θ) are formed by forcing air through the
teeth. The apex of the tongue is normally inserted between the upper and the
lower teeth and air is forced out. This causes a friction which is realized in the
form of two different consonant sounds: /ð/ as in this /ðɪs/, and /θ/ as in thin
/θɪn/. If you say the soft /θ/ in /thin/ and then the hard /ð/ sound in /then/, you

48 CHAPTER THREE
can feel the air being forced through the teeth. The tongue tip and rims are
articulating with the upper teeth. These sounds are called dental or interdental
sounds. The upper teeth are also used when you say /f/ and /v/. In this case
however, air is being forced through the upper teeth and lower lip.

2.2.3. ALVEOLAR RIDGE

An alveolar sound is produced when the tongue tip, or blade, touches the bony
prominence behind the top teeth. This prominence is in fact that part of the
gum which lies behind the upper teeth. In English, the following phonemes are
alveolar: /t/ as in tin /tɪn/, /d/ as in din /dɪn/, /s/ as in sin /sɪn/, /z/ as in zip /zɪp/,
/l/ as in look /lʊk/, /r/ as in roof /ru:f/, and /n/ as in /night /naɪt/. Try saying all of
these phonemes to yourself. They do not all touch in exactly the same way due
to the manner of articulation. Some are plosives while the others are fricatives
or laterals. But the place of articulation is clearly the alveolar ridge for all of
them. The Iranian EFL learners' attention is specifically drawn to the difference
between the Persian and English /t/ and /d/ phonemes. Whereas in English
these sounds are purely alveolar stops, in Persian they are dentoalveolar
phonemes. The blade of the tongue touches the back of upper teeth to form
these sounds in Persian whereas, in English, the tongue lies solely behind the
alveolar ridge. The difference can be felt if English /t/ and /d/ phonemes are
inserted in Persian words.

PHONEME EXAMPLE PRONUNCIATION

/ʃ/ sheep /ʃi:p/
/ʒ/ genre /ˈʒɑ:nrə/
/ʧ/ cheap /ʧi:p/
/ʤ/ jeep /ʤi:p/

Four sounds are said to be palatoalveolar. This is partly because the blade of
the tongue straddles both the alveolar ridge and the front of the hard palate as

CHAPTER THREE 49
air is forced through to make the sounds. These sounds include /ʃ/, /ʒ/, /ʧ/, and
/ʤ/. Take the following examples:

2.2.4. HARD PALATE

This is the hard bit of the arched bony structure that forms the roof of the
mouth. The /j/ sound in yes is the clearest example of a palatal sound in RP.
You can feel the fricative sound being forced between the tongue and the very
top of your mouth. The blade of the tongue touches the hard palate to articulate
palatal sounds. Earliest accounts of phonetics also classified affricates as
being palatal. Today, however, they are considered to be palatoalveolar rather
than palatal per se. Still, there are some phoneticians who prefer the older
classification.

2.2.5. THE SOFT PALATE (VELUM)

The soft palate is toward the back of the mouth. It is where the roof of the
mouth gives way to the soft area behind it. It can just be felt with your tongue if
you curl it as far back and as high as you can so that the apex of your tongue
can feel the soft area of the back-roof of your mouth. The soft palate is
technically called the velum and sounds which are produced by a constriction
or blockage at this part of the vocal tract are called velar sounds. Thus, velar
sounds are usually made when the back of the tongue is pressed against the
soft palate. They include the /k/ in cat, the /g/ in girl and the /ŋ/ in hang. The
glide /w/ is also regarded as a labiovelar sound, because it simultaneously
uses both lips whilst raising the back of the tongue towards the velum. Try
saying /wheel/ and /win/ and feel the position of your tongue.

2.2.6. GLOTTIS

Glottal sounds are those sounds that are made in the larynx through the
closure or narrowing of the glottis. It was mentioned earlier that there is an
opening between the vocal cords which is called the glottis. If pulmonary air is
constricted or blocked at this place, glottal sounds are articulated. The glide /h/

50 CHAPTER THREE
as in Helen is an example of a glottal sound. It is physically impossible to feel
the process using your tongue. It is as far back as you can get in your mouth.
Unlike English, Persian syllables cannot start with a vowel. As such, Iranian
EFL learners are inclined to insert the glottal stop consonant /ʔ/ at the
beginning of English syllables that start with a vowel. The /æ/ sound, as it
appears in Arabic, is an extreme example of a glottal stop. The glottal stop is
becoming a more widespread part of British English, but is still uncommon in
RP. You also use your glottis for speech when you whisper or speak in a
creaky voice. Take the following examples:

WORD ENGLISH IRANIAN MISPRONUNCIATION

ago /əˈgəʊ/ /ʔəˈgəʊ/
out /aʊt/ /ʔaʊt/
apple /æpl/ /ʔæpl/

3. CONSONANTS

Consonants in English are distinguished from vowels on the basis of the

modifications of pulmonary air in the oral cavity. Consonants are distinguished
from one another on the basis of their differences in three respects: (a) manner
of articulations, (b) place of articulation, and (c) voicing. By way of contrast,
vowels are distinguished from one another on the basis of two criteria: (a)
relative position of the tongue in the mouth, and (b) lip rounding.

Consonant is the general term that refers to a class of sounds where there is
obstruction of some kind (i.e., complete blockage, or constriction) to the flow of
pulmonary air. As it was mentioned earlier, there are six different degrees of
obstruction. Therefore, consonants can be classified into six different
categories on the basis of their manner of articulation:

CHAPTER THREE 51
 TYPE PHONEME
PLOSIVES /p/ /b/ /t/ /d/ /k/ /g/
FRICATIVES /f/ /v/ /θ/ /ð/ /h/ /s/ /z/ /ʃ/ /ʒ/
AFFRICATES /ʧ/ /ʤ/
NASALS /m/ /n/ /ŋ/
APPROXIMANT /r/ /w/ /j/
LATERAL /l/

However, as the table shows, more than one consonants fall within almost all
of these categories. Therefore, other criteria are needed to distinguish one
consonant from the other. For example, /p/ and /b/ cannot solely be
distinguished on the basis of their manner of articulation. Moreover, they are
articulated at the same place of articulation. Yet they are different since they
assign different meanings to the two English words 'pat' /pæt/ and 'bat' /bæt/.

Consonants that share the same manner of articulation may be different in

terms of place of articulation. Consonants are classified into nine different
classes according to their place of articulation:

TYPE PHONEME
BILABIAL /m/, /p/, /b/
LABIODENTAL /f/, /v/
INTERDENTAL /θ/, /ð/
ALVEOLAR /t/, /d/, /n/, /s/, /z/, /r/, /l/
PALATOALVEOLAR /ʃ/, /ʒ/, /ʧ/, /ʤ/
PALATAL /j/
LABIOVELAR /w/
VELAR /k/, /g/, /ŋ/
GLOTTAL /h/

52 CHAPTER THREE
Even those consonant that share the place and manner of articulation may be
different in terms of voicing and nasality. According to the level of vibration of
the vocal cords, consonants are classified into two groups: voiced, and
voiceless:

TYPE PHONEME
VOICELESS /ʧ/, /s/, /p/, /k/, /f/, /ʃ/, /t/, /θ/, /h/
VOICED /ʤ/, /z/, /b/, /g/, /v/, /ʒ/, /d/, /ð/, /w/, /j/, /l/, /r/, /m/, /n/,
/ŋ/

On the basis of nasality, consonants are divided into two groups: nasal, and
non-nasal. As you have already noticed, nasality is identified by a free flow of
air through the nose.

TYPE PHONEME
NASAL /m/, /n/, /ŋ/
NON-NASAL /ʤ/, /z/, /b/, /g/, /v/, /ʒ/, /d/, /ð/, /w/, /j/, /l/, /r/, /ʧ/,
/s/, /p/, /k/, /f/, /ʃ/, /t/, /θ/, /h/

These differences in place of articulation, manner of articulation, nasality, and

voicing led traditional phoneticians to assign different names to different
consonants. The name that was given to any given consonant was based on
the air stream mechanism which led to the articulation of that consonant:

Consonant name = Place of articulation + voicing + manner of articulation

For example the consonant /p/ would be identified as bilabial voiceless stop. By
way of contrast, the consonant /b/ was defined as bilabial voiced stop. As such
/b/ and /p/ were distinguished on the basis of the level of vibration of the vocal
cords (i.e., voicing). /m/ was considered to be a bilabial nasal.

In traditional phonetics, consonants were named after their particular

characteristics:

CHAPTER THREE 53
 PHONEME TRADITIONAL NAME
/p/ Bilabial voiceless stop
/b/ Bilabial voiced stop
/m/ Bilabial nasal
/f/ Labiodental voiceless fricative
/v/ Labiodental voiced fricative
/θ/ Interdental voiceless fricative
/ð/ Interdental voiced fricative
/t/ Alveolar voiceless stop
/d/ Alveolar voiced stop
/n/ Alveolar nasal
/s/ Alveolar voiceless fricative
/z/ Alveolar voiced fricative
/l/ lateral
/r/ Alveolar approximant
/ʃ/ Palatoalveolar voiceless fricative
/ʒ/ Palatoalveolar voiced fricative
/ʧ/ Voiceless affricate
/ʤ/ Voiced affricate
/j/ Palatal approximant
/w/ Labiovelar approximant
/k/ Velar voiceless stop
/g/ Velar voiced stop
/ŋ/ Velar nasal
/h/ Glottal fricative

Also notice that earlier phoneticians had a tendency to distinguish

approximants, glides, and laterals as fricatives. In addition, they also classified
/s/, /z/, /ʃ/, /ʒ/, /ʧ/, /ʤ/, /r/, and /l/ as palatals.

54 CHAPTER THREE
A closer look at the names given to consonants reveals that only the minimum
number of criteria needed to distinguish one consonant from the rest should be
used to name that consonant. These features are called distinctive features. In
addition to distinctive features, there are a number of characteristics of
phonemes that can be guessed on the basis of their distinctive features. These
predictable characteristics are called redundant features. Redundant features
should not be mentioned in the naming of phonemes. The phoneme /ŋ/, for
example, is voiced. However, the voicing feature has not been used in its
name. This is because it is possible to guess that the phoneme /ŋ/ is voiced
due to the fact that it is a nasal consonant. In other words, nasality can predict
voicing. Therefore, voicing is said to be a redundant feature for nasal
consonants, and is not used in their names. A feature which is redundant for
one phoneme may be distinctive for another. For example, whereas voicing is
considered to be a redundant feature for the phoneme /ŋ/, it is certainly a
distinctive feature for the phonemes /g/, and /k/ because voicing is the only
difference between them.

Phoneticians usually express redundant features in terms of statements which

they call redundancy rules. Thus, the fact that voicing can be predicted by
nasality is expressed in the following redundancy rule: whenever a phoneme is
nasal, it is deemed to be voiced. Like other scientific fields, phonetics draws on
a system of notations to simplify its redundancy rules. Take the following
example: [nasal] → [voiced] (where → is read as 'rewrites as'). Redundancy
rules will be fully discussed in the following chapters.

4. VOWELS

It was mentioned earlier that most sounds in speech are produced by passing a
stream of air from the lungs through one or more resonators belonging to the
phonetic apparatus. There are four principle resonators: (1) the pharyngeal
cavity, (2) the oral cavity, (3) the labial cavity, and (4) the nasal cavity. Air flows
through these resonators.

CHAPTER THREE 55
 RESONATORS:
1) LABIAL
2) ORAL
3) NASAL
4) PHARYNGEAL

The presence or absence of obstructions in the course of the airstream

modifies the nature of the sound produced. By classifying the different types of
obstructions that are possible, articulatory phonetics distinguishes four different
major sound classes: consonants, vowels, glides or semivowels, and liquids.

The distinction between consonants and vowels is quite simple. If the air, once
out of the glottis, is allowed to pass freely through the oral cavity, the sound is a
vowel. If the air, once out of the glottis, is partially or totally obstructed in one or
more places in the oral cavity, the sound is a consonant. It should be noted that
the line between vowels and consonants cannot be clearly drawn. A continuum
exists between the two extremes. In English, there are also intermediate
instances: (a) liquids, and (b) glides.

In order for a phoneme to be a vowel, it should meet certain vowel-hood

criteria. These criteria include: (a) the degree of openness of the oral cavity

56 CHAPTER THREE
also known as the degree of aperture, (b) the degree of tension or laxity of the
vocal tract muscles, and (c) amount of duration or length of articulation.

There is a large degree of freedom in the articulation of open vowels. However,

this freedom is not endless. On the one hand, no vowel can be more 'open'
than the standard open vowels (fourth degree of aperture). On the other hand,
a vowel could not be much more 'close' than the standard close vowels (first
degree of aperture). Additionally, the close vowels must have a certain
minimum duration in order to be perceived as a vowel rather than a consonant.

The chief characteristic of vowels is the freedom with which the airstream, once
out of the glottis, passes through the speech organs. The supra-glottal
resonators do not cut off or constrict the air. They only cause resonance, that is
to say, the reinforcement of certain frequency ranges.

A vowel's timbre (or quality) depends on the following elements: (a) the number
of active resonators, (b) the shape of the oral cavity, and (c) the size of the oral
cavity. There are three possible resonators involved in the articulation of a
vowel: (a) the oral cavity, (b) the labial cavity, and (c) the nasal cavity. If the
soft palate is raised, the air does not enter the nasal cavity, and passes
exclusively through the oral cavity; if the soft palate is lowered, however, air
can pass through nose and mouth simultaneously. If the lips are pushed
forward and rounded, a third, labial resonator is formed; if, on the other hand,
the lips are spread sideways or pressed against the teeth, no labial resonator is
formed. It is thus the number of resonators at stake in distinguishing nasal
vowels (nasal resonator active) from oral vowels (no nasal resonance), and
rounded vowels (labial resonator active) from unrounded vowels (no labial
resonator/no labial resonance).

Ordinarily, English vowels do not involve the nasal cavity. They can, however,
become nasalized in certain contexts (e.g., when they follow nasal
consonants). The shape of the oral cavity is determined by the general position
of the tongue in the mouth. This divides the vowels into three great classes: (a)

CHAPTER THREE 57
front vowels, (b) back vowels, and (c) central vowels. In the articulation of front
vowels, the tongue body is in the pre-palatal region. In the production of back
vowels, the tongue body is in the post-palatal or velar region. Finally, in the
articulation of central vowels, the tongue body is in the medio-palatal region.

FRONT VOWELS BACK VOWELS

The size of the oral cavity is the last factor considered in the articulatory
classification of a vowel. The size of the oral cavity depends directly upon the
degree of aperture (of the mouth), that is, upon the distance between the
palate and the tongue's highest point. Arbitrarily, four degrees of aperture are
distinguished, from the most closed (first degree) to the most open (fourth
degree). The descriptions of the vocalic articulations (i.e., names given to
vowels) should be grouped according to the following principle:

An initial classification is made based on degree of aperture; within each group,

the vowels are then divided according to mouth shape, and then as rounded or
unrounded.

58 CHAPTER THREE
 CLOSE VOWELS HALF-CLOSE VOWELS

HALF-OPEN VOWELS OPEN VOWELS

Vowels differ from consonants in that there is no noticeable obstruction in the

vocal tract during their production. Air escapes in a relatively unimpeded way
through the mouth and/or nose. If you try saying aaaaa, iiiii, uuuuu, eeeeee,
oooooo to yourself you should be able to feel that, although your tongue moves
about your mouth, it never actually obstructs the airflow.

CHAPTER THREE 59
 Tongue positions for vowel production (Reprinted from Lieberman and Blumstein,
1988, P. 126)

You should also be able to feel that the position of the tongue changes for each
of those vowels. Vowels are far more difficult to transcribe than consonants and
are also an extremely important area of English phonology as they make up the
greatest difference between English varieties. In other words, the differences
among varieties of English are mostly due to the speakers' articulation of
vowels. In addition, vowels are the stress carrying phonemes in stressed
syllables.

The relative position of the tongue in the mouth is the most crucial criterion that
determines the differences between vowels. The changes of the tongue
position can be very slight and difficult to detect. Therefore, phoneticians
produced what they called the vowel chart to describe these tongue positions.

60 CHAPTER THREE
As such, the vowel chart attempts to map the positions of the tongue and jaw in
articulating vowels. It is worth mentioning that, as it is so difficult to determine
the exact position of the lips, tongue and palate, there is no single agreed upon
vowel chart.

4.1. THE VOWEL CHART

The vowel chart is quite a complicated diagram.

All that the vowel chart is trying to do is to represent where the tongue lies in
relation to the openness of the mouth when you sound a vowel. So the front
closed vowel /i:/ means that your tongue is in a forward position in the mouth
which is in a relatively closed position. Try saying it to yourself and then
contrast it with the open back sound in the diagram.

CHAPTER THREE 61
As such, the English vowels can be represented in the form of a table or chart
that can show the relative position of the tongue in the mouth. This chart can
tell the location in the mouth where a vowel is articulated. It can also indicate
the degree to which the mouth is open (technically known as degree of
aperture) when a given vowel is produced. For example, you may have the
experience of going to a doctor because of a cold or flu. The doctor wants you
to say æææææææ to be able to see the back of your mouth. This shows that
the /æ/ sound is articulated with a lowered jaw which adds to the openness of
the mouth. Look at the following figure to see how English vowels are
articulated.

4.2. TYPOLOGY OF VOWELS

There are two major types of vowels on the basis of their length: long vowels,
and short vowels. Long vowels are usually distinguished from short vowels in
that the duration of time that speakers spend in articulating them is somewhat
longer than the time spent for the articulation of short vowels. This
characteristic of long vowels is known as length and is schematically
represented by putting a colon (:) after the phoneme which represents the
vowel in question. In British English, there are five long vowels:

/i:/ as in sheep /ʃi:p/

/ɜ:/ as in fur /fɜ:/
/ɔ:/ as in four /fɔ:/
/ɑ:/ as in car /kɑ:/
/u:/ as in boot /bu:t/

As it was mentioned above, short vowels are formed in a much shorter time
than long vowels. In other words, short vowels lack the length feature. In British
English, there are seven short vowels:

/ɪ/ as in hit /hɪt/

/e/ as in hen /hen/

62 CHAPTER THREE
 /æ/ as in hat /hæt/
/ə/ as in ago /ə’gəʊ/
/ʊ/ as in book /bʊk/
/ʌ/ as in bus /bʌs/
/ɒ/ as in hot /hɒt/

In English, vowels can also glide into one another to form diphthongs (i.e.,
sequence of two vowels realized as one sound in pronunciation) and even
triphthongs (i.e., sequence of three vowels realized as one sound in
pronunciation).

Diphthongs are those sounds that consist of a movement or glide from one
vowel to another. The first part of a diphthong is always longer and stronger
than the second part; as the sound glides into the second part of the diphthong
the loudness of the sound decreases. One of the most frequent errors made by
language learners is to use pure vowels instead of diphthongs. English
diphthongs include the following:

/ʊə/ as in poor /pʊə/

/ɔɪ/ as in boy /bɔɪ/
/ɪə/ as in here /hɪə/
/əʊ/ as in know /nəʊ/
/eɪ/ as in day /deɪ/
/aʊ/ as in cow /kaʊ/
/aɪ/ or /ʌɪ/ as in my /maɪ/ or /mʌɪ/
/eə/ or /ɛə/ as in hair /heə/ or /hɛə/

Triphthongs are those sounds that consist of a movement or glide from one
vowel to another and then onto a third. They are very similar to diphthongs, but
have an extra schwa on the end of the diphthongs. There are said to be only
five triphthongs, but there are a number of occasions when diphthongs meet

CHAPTER THREE 63
other vowels over word edge boundaries. The five clear examples of
triphthongs are as follows:

/aʊə/ as in power /paʊə/

/aɪə/ as in liar /laɪə/
/eɪǝ/ as in layer /leɪǝ/
/ɔɪǝ/ as in loyal /lɔɪǝl/
/ǝʊǝ/ as in mower /mǝʊǝ/

4.3. WEAK FORMS OF VOWELS

Vowels in English can appear in stressed and unstressed syllables. When a

vowel appears in a stressed syllable, it is pronounced with its strong or full
form. However, when vowels appear in unstressed syllables, they are normally
neutralized/centralized and adopt a rather weak identity known as the weak
form. The weak form of all vowels is /ə/ except for /i:/ and /ɪ/ of which the weak
form is /ɪ/. This can be schematically represented as the following:

Strong /i:/ /ɪ/ /u:/ /ʊ/ /ɔ:/ /ɒ/ /ɑ:/ /ʌ/ /ɜ:/ /æ/ /e/
Weak /ɪ/ /ɪ/ /ə/ /ə/ /ə/ /ə/ /ə/ /ə/ /ə/ /ə/ /ə/

In morphology, weak forms are those words that are pronounced in an

unstressed manner. Many of the most common words in English can come in
either a strong or a weak form. The most crucial factors in determining whether
a word should be pronounced with a weak vowel or a strong one are (a) the
type of the word (i.e., function words versus content words), and (b) the
number of syllables that compose the word (i.e., whether the word is mono-
syllabic, bi-syllabic, or poly-syllabic). A function word is a word that has a
grammatical rather than semantic function in the language. In other words, the
words which are used to express grammatical relationships among words in a
sentence are called function words. Functions words include: conjunctions,
articles, pronouns, prepositions and some auxiliary and modal verbs. Content

64 CHAPTER THREE
words, on the other hand, are rather encyclopedic than linguistic. That is, those
words that have independent encyclopedic meanings of their own, and that can
be used independently, are called content words. Content words include: main
verbs, adjectives, adverbs, and nouns.

THE /ðə/ before consonants (the man /ðəmæn/)

ARTICLES

/ði/ before vowels (the orange /ði’ɒrɪnʤ/)

A, AN /ə/ before consonants (a bike /əbaɪk/)
/ən/ before vowels (an apple /ənæpl/)
AND /ən(d)/ (come and see /kɒmənsi:/)
CONJUNCTIONS

Fish and chips /fiʃənʧɪps/

BUT /bət/ (cheap but reliable /ʧI:pbətrɪlaɪəb(ə)l/)
(Last but one /læstbətwʌn/)
THAN /ðən/ (better than expected /betəðənɪk’spektɪd/)
(More than one /mɔ:ðənwʌn/)
HIS /hɪz/, /həz/, /əz/ before a noun
PRONOUNS

Take his coat /teɪkəzkəʊt/

HER /ə/ before consonants (take her home /teɪkəhəʊm/)
/ə(r)/ Before vowels (take her out /teɪkə(r)aʊt/)
un-emphatically or in non-sentence-final positions
PREPOSITIONS

FROM /frəm/ (he is from london /hi:ɪzfrəmlɒnd(ə)n/)

AT /ət/ (at the table /ətðəteɪbl/)
OF /əv/ (the leg of the table /ðəlegəvðəteɪbl/)
ON /ən/ (and on thy cheeks /ənənðaɪʧi:ks/)
etc.

Weak forms are all function words. Generally the strong forms of these words
are used (a) when they are being directly quoted, (b) when they are being
contrasted or (c) if they appear at the end of a sentence. The pronunciation of a
weak form can be so different from the strong form that it is barely recognizable

CHAPTER THREE 65
as being the same word. If said in isolation, it would be all but unintelligible.
Usually, it is the context that makes it understandable.

It is possible to use only strong forms of vowels in English, and some non-
native speakers do exactly this. However, it sounds very unnatural to a native
speaker and it will also mean that a person who only uses the strong forms of
English vowels will have trouble understanding native speakers of English who
use the weak form all of the time. The most common weak form examples are
introduced as follows:

In traditional phonetics, vowels were named on the basis of such criteria as the
relative position of the tongue in the mouth, lip rounding, and length.

VOWEL NAME VOWEL NAME

/i:/ High front vowel /ʌ/ Low mid vowel
/ɪ/ High-mid front vowel /u:/ High back vowel
/e/ Mid front vowel /ʊ/ High-mid back vowel
/æ/ Low front vowel /ɔ:/ Mid back vowel
/ǝ/ Mid mid vowel /ɒ/ Mid-low back vowel
/ɜ:/ Mid-low mid vowel /ɑ:/ Low back vowel

As such, the vowel /i:/ would, for example, be considered a high front vowel.
Other vowels also received their names in this way:

vertical position + horizontal position + vowel.

5. SEMIVOWELS OR GLIDES

Semivowels or glides are those sounds which, though articulated quite

variously, in one way or another fail to meet the above criteria for full vowel-
hood. They also fail to meet the obstruction or blockage criterion for full
consonant-hood. They are, therefore, considered to be an intermediate state
between vowels and consonants. They involve the same restriction of the
speech canal as fricative consonants, but the speech organs are substantially

66 CHAPTER THREE
less tense during the articulation of semivowels. Rather than friction, a
resonant sound is produced at the place of articulation. The English language
includes three semivowels: /h/, /w/, and /j/.

6. LIQUIDS

Liquids meet not only the vowel-hood criteria but also the consonant-hood
criterion. They are, like semivowels, considered to be an intermediate state
between vowels and consonants. Liquids are generally considered to be a
special case. During their production, the oral cavity is constricted while the
tongue sinks down in such a way as to channel the air out of the mouth as if a
vowel is being articulated. English includes two liquid sounds: /l/, and /r/.

EXERCISE

1. Look at the following illustrations:

1 2 3 4

5 6 7 8

CHAPTER THREE 67
Each of these illustrations represents the shape of the human vocal organ for
the articulation of a sound or a class of sounds. Complete the following table by
indicating the sound/class of sounds represented by each illustration.

ILLUSTRATION YOUR ANSWER

e.g., 4 Front vowels

2. Write the symbol that corresponds to each of the following phonetic

descriptions, then give an English word that contains this sound.

Example: voiced alveolar stop [ d ] dough

A voiceless bilabial unaspirated stop [ ]
B low front vowel [ ]
C lateral liquid [ ]
D velar nasal [ ]
E voiced interdental fricative [ ]
F voiceless affricate [ ]
G palatal glide [ ]
H mid lax front vowel [ ]
I high back tense vowel [ ]
J voiceless aspirated alveolar stop [ ]

68 CHAPTER THREE
3. For each group of sounds listed, state the phonetic feature(s) they all share.

Example: [p] [b] [m] Features: bilabial, stop, consonant

GROUP OF SOUNDS SHARED FEATURES

A [g] [p] [t] [d] [k] [b]

B [u:] [ʊ] [ɔ:] [ɒ]

C [i:] [ɪ] [ɜ:] [e] [æ]

D [t] [s] [ʃ] [p] [k] [ʧ] [f] [h] [θ]

E [v] [z] [ʒ] [ʤ] [n] [g] [d] [b] [l] [r] [w] [j]

F [θ] [ð]

G [t] [d] [n] [s] [z] [r] [l]

H [ʃ] [ʒ] [ʧ] [ʤ]

I [k] [g] [ŋ]

J [j] [w] [h]

K [t] [d] [s] [ʃ] [n] [ʧ] [ʤ]

4. What phonetic property or feature distinguishes the sets of sounds in column

A from those in column B?

COLUMN A COLUMN B YOUR ANSWER

A [i:] [ɪ] [u:] [ʊ]
B [p] [t] [k] [s] [f] [b] [d] [g] [z] [v]
C [p] [b] [m] [t] [d] [n] [k] [g] [ŋ]
D [i:] [ɪ] [u:] [ʊ] [e] [ɜ:] [ɔ:] [ɒ] [æ] [ʌ]
E [f] [v] [s] [z] [ʃ] [ʒ] [ʧ] [ʤ]
F [i:] [ɪ] [e] [ə] [ɜ:] [æ] [u:] [ʊ] [ɔ:] [ɒ] [a:]

CHAPTER THREE 69
5. Which of the following sound pairs have the same manner of articulation?
Mark Ⓨ if the sound pairs have the same manner of articulation, and Ⓝ if
they have different manners of articulation.

A [h] [ʔ] Ⓨ Ⓝ F [f] [ʃ] Ⓨ Ⓝ

B [r] [w] Ⓨ Ⓝ G [k] [θ] Ⓨ Ⓝ

C [m] [ŋ] Ⓨ Ⓝ H [s] [g] Ⓨ Ⓝ

D [ð] [v] Ⓨ Ⓝ I [j] [w] Ⓨ Ⓝ

E [t] [r] Ⓨ Ⓝ J [j] [ʤ] Ⓨ Ⓝ

6. Identify the manner of articulation for each of the following sounds.

YOUR ANSWER YOUR ANSWER

[h] ( ) [f] ( )

[ʔ] ( ) [ʃ] ( )

[w] ( ) [k] ( )

[ŋ] ( ) [ʤ] ( )

[ð] ( ) [ʧ] ( )

[θ] ( ) [ʒ] ( )

[v] ( ) [j] ( )

[r] ( ) [s] ( )

[t] ( ) [z] ( )

[m] ( ) [n] ( )

70 CHAPTER THREE
 CHAPTER FOUR
SYSTEMATIC ARTICULATORY PHONETICS

1. INTRODUCTION

Despite its potentials, traditional phonetics was inadequate in a number of

cases. For example, the phoneme /p/, as it was mentioned earlier, has two
allophonic realizations [p], and [ph]. Vowels, too, can mix to form diphthongs or
triphthongs. However, the naming convention used in traditional phonetics
could not easily capture these nuances. Therefore, phoneticians sought to
develop a new comprehensive and exhaustive system. This new perspective
on phonetics was called systematic or modern phonetics.

2. WHAT IS SYSTEMATIC/MODERN PHONETICS

Modern phonetics is based on the idea that each phonetic segment or

phoneme is composed of a bundle of features. These features provide a good
description of the phoneme. As such, phonemes are distinguished from one
another on the basis of their phonetic features. The presence of a feature in the
bundle of features that identify a phoneme is marked by the plus [+] symbol
and the absence of the feature by the minus [-] symbol. Therefore, the phonetic
features used in modern phonetics are called binary features due to their dual
[±] values. For example, the allophone [ph] is said to be [+aspiration] whereas
its allophone [p] is identified as [-aspiration]. The symbol [α] is used to show
both [±] values. The symbol [-α] is used to show the [∓] value.

Each phoneme is, therefore, adequately represented as a matrix of binary

phonetic features. All sounds marked by the same value [+ or -] for a feature
belong to a natural class named after that feature. For instance, all sounds that
are marked [+voice] belong in the voiced class and all sounds that are marked

CHAPTER FOUR 71
[-voice] belong in the voiceless class. Unlike traditional phonetics, phonemes
are not given individual names in modern or systematic phonetics. Rather, the
bundle of distinctive features that distinguish one phoneme from the rest is
used as a description (or name) for that phoneme. For example, the nasal
consonants /ŋ/, /n/, and /m/ are described as:

/ŋ/ /n/ /m/

-vocalic -vocalic -vocalic
+consonantal +consonantal +consonantal
+nasal +nasal -coronal
-anterior +coronal +anterior

Each of these features describes one of the discrete activities in (or parts of)
the vocal tract that are crucial to the articulation of the phoneme. It should be
noted that, like in traditional phonetics, in modern phonetics, too, each
phoneme may be composed of more features than are actually in its
description. However, many features are predictable on the basis of other
features. In our example here, the phoneme /ŋ/ is a [+voice] phoneme.
However, we did not use this feature in the bundle of features that describe the
phoneme because the feature [+nasal] can predict the feature [+voice].
Therefore, only the minimum inevitable number of features that differentiate
between a given phoneme and the rest of sound segments should be used in
the description of that phoneme. These are, thus, called the distinctive features
of that phoneme. The predictable features of a phoneme are called the
redundant features of that phoneme.

Phonetic features not only differentiate between phonemes but also between
classes of phonemes. In fact the notion of natural classes of sounds derives
from this ability of binary features. For instance, /m/, /n/, and /ŋ/ are all [+nasal]
while other phonemes are [-nasal]. So, these three phonemes compose a
natural class of sounds in English. A natural class is defined as: A class of

72 CHAPTER FOUR
phonemes in which the number of features that must be specified to define that
class is smaller than the number of features required to distinguish any
member of that class.

In our example, only the feature [+nasal] is required to distinguish the nasal
class whereas, as you have noticed, each of the nasal phonemes /ŋ/, /n/, and
/m/ needs more than three distinctive features to be identified. Nasals do not
comprise the only natural class of sounds. There are, in fact, many different
natural classes of sounds in the English language. The major natural classes of
sounds of English are consonants, vowels, liquids, and glides. These classes
can easily be distinguished from each other on the basis of values (±) they
assign to the phonetic features [consonantal] and [vocalic]. Compare the
following:

VOWELS CONSONANTS
+vocalic -vocalic
-consonantal +consonantal

LIQUIDS GLIDES
+vocalic -vocalic
+consonantal -consonantal

Before we embark on any discussion of phonetic features, we had better

identify English phonemes in terms of their distinctive features.

1) English liquids are identified by the following features:

/r/ /l/
+consonantal +consonantal
+vocalic +vocalic
-lateral +lateral

CHAPTER FOUR 73
 2) English glides are identified by the following features:

/j/ /w/ /h/

-vocalic -vocalic -vocalic
-consonantal -consonantal -consonantal
+voice +voice -voice
-round +round

3) English voiceless consonants are identified by the following features:

/p/ /f/ /θ/

-vocalic -vocalic -vocalic
+consonantal +consonantal +consonantal
-voice -voice -voice
+abrupt release +strident +continuant
+anterior -coronal -strident
-coronal +coronal
/t/ /s/ /ʃ/
+consonantal +consonantal +consonantal
-vocalic -vocalic -vocalic
-voice -voice -voice
+abrupt release +strident +strident
+coronal +anterior +continuant
+coronal +coronal
-anterior
/ʧ/ /k/
+consonantal +consonantal
-vocalic -vocalic
-voice -voice
-abrupt release +abrupt release
-continuant -anterior

74 CHAPTER FOUR
 4) English voiced consonants are identified by the following features:

/b/ /v/ /ð/

-vocalic -vocalic -vocalic
+consonantal +consonantal +consonantal
+voice +voice +voice
+abrupt release +strident +continuant
+anterior -coronal -strident
-coronal +coronal
/d/ /z/ /ʒ/
+consonantal +consonantal +consonantal
-vocalic -vocalic -vocalic
+voice +voice +voice
-nasal +strident +strident
+abrupt release +anterior +continuant
+coronal +coronal +coronal
-anterior
/ʤ/ /g/
+consonantal +consonantal
-vocalic -vocalic
+voice +voice
-abrupt release -nasal
-continuant +abrupt release
-anterior

5) English nasals are identified by the following features:

/ŋ/ /n/ /m/

-vocalic -vocalic -vocalic
+consonantal +consonantal +consonantal
+nasal +nasal -coronal
-anterior +coronal +anterior

CHAPTER FOUR 75
Like other English phonemes, English vowels, too, have been identified in
terms of distinctive features. However, the set of distinctive features used to
identify vowels is somewhat different from the set of features used for the
identification of other phonemes. English vowels are identified in terms of the
following features:

/i:/ /ɪ/ /e/

-consonantal -consonantal -consonantal
+vocalic +vocalic +vocalic
+high +high -high
-back -back -low
+tense -tense -back
+tense

/ɜ:/ /æ/ /ə/

-consonantal -consonantal -consonantal
+vocalic +vocalic +vocalic
-high +low -low
-low -back -high
-back -back
-tense -tense

/ʌ/ /ɑ:/ /ɒ/

-consonantal -consonantal -consonantal
+vocalic +vocalic +vocalic
-high +low -high
-low +back -low
-back +back
-rounded -tense
+rounded

76 CHAPTER FOUR
 /ɔ:/ /u:/ /ʊ/
-consonantal -consonantal -consonantal
+vocalic +vocalic +vocalic
-high +high +high
-low +back +back
+back +rounded +rounded
+tense +tense -tense
+rounded

The features used to distinguish English vowels can be summed up in the form
of a matrix (table 4.1. below).

/ i:/ /ɪ/ /e/ /ɜ:/ /æ/ /ʌ/ /u:/ /ʊ/ /ɔ:/ /ɒ/ /ɑ:/
high + + - - - - + + - - -
Mid - - + + - + - - + + -
low - - - - + - - - - - +
back - - - - - - + + + + +
Central - - - - - + - - - - -
rounded - - - - - - + + + + -
tense + - + - - - + - + - -
Table 4.1.: Phonetic features of English vowels (Reprinted from Fromkin, Rodman,
and Hyams (2003) p. 299—with some modifications).

Notice that all English vowels are [+syllabic] and [+voiced]. Also notice that the
identification of diphthongs and triphthongs requires more than one column of
features. The diphthongs would be represented by a two-column feature matrix
of the vowel followed by the glide; the triphthongs would be represented by a
three-column feature matrix of the vowel followed by the glide followed by the
vowel.

CHAPTER FOUR 77
 Abrupt release
PHONEME

Consonantal

Continuant
Aspiration

Sonorant
Anterior
Coronal
Strident

Syllabic
Vocalic

Lateral
Nasal
Voice
PLOSIVES /p/ - + - - + - - + - - - - -
/b/ - + + - + - - + - - - - -
/t/ - + - - + - - + + - - - -
/d/ - + + - + - - + + - - - -
/k/ - + - - + - - - - - - - -
STOPS

/g/ - + + - + - - - - - - - -
NASALS /m/ - + + + + - - + - - - + ±
/n/ - + + + + - - + + - - + ±
/ŋ/ - + + + + - - - - - - + ±
AFFRICATES /ʧ/ - + - - - - - - + + - - -
/ʤ/ - + + - - - - - + + - - -
FRICATIVES /f/ - + - - - + - + - + - - -
/v/ - + + - - + - + - + - - -
/θ/ - + - - - + - + + - - - -
/ð/ - + + - - + - + + - - - -
/s/ - + - - - + - + + + - - -
/z/ - + + - - + - + + + - - -
/ʃ/ - + - - - + - - + + - - -
/ʒ/ - + + - - + - - + + - - -
TABLE 4.2.: Features of English consonants.

The phonetic features that can be used for the identification of English
consonants have been summarized in table 4.2. above.

The phonetic features of English glides and liquids are identified in table 4.3.
below.

78 CHAPTER FOUR
 Abrupt release
PHONEME

Consonantal

Continuant
Aspiration

Sonorant
Anterior
Coronal
Strident

Syllabic
Vocalic

Lateral
Nasal
Voice
LIQUIDS /l/ + + + - - + - + - - + + ±
/r/ + + + - - + - + - - - + ±
GLIDES /h/ - - - - - + - - - - - + -
/w/ - - + - - + - - - - - + -
/j/ - - + - - + - - - - - + -
TABLE 4.3.: Features of English liquids and glides.

3. AIRSTREAM MECHANISMS

It was pointed out before that most sounds in speech are produced by passing
a stream of air from the lungs through one or more resonators belonging to the
phonetic apparatus. In a small number of languages, the airstream does not
originate in the lungs, but rather from outside. This velaric ingressive airstream
mechanism produces implosive and click sounds through inhalation. A speech
sound can also be generated from a difference in pressure of the air inside and
outside a resonator. In the case of pharyngeal cavity, this pressure difference
can be created without using the lungs at all whereby producing ejective
sounds. This is called the glottalic egressive airstream mechanism. The main
airstream mechanism in most of the world's languages, however, is the
pulmonic egressive airstream mechanism which produces almost all the
phonemes of the world's most important languages. In this case, the flow of air
originates in the lungs. As such, it is called pulmonic airstream (or simply
pulmonic air). It then moves outward through the resonators; hence, the name
egressive. When articulators (i.e., parts of the vocal tract that modify the
airstream) influence the flow of pulmonic air, sounds are produced. The nature
of the modifications that produce one phoneme differs from those of any other

CHAPTER FOUR 79
phoneme. As it was mentioned earlier, modern or systematic phonetics uses
binary phonetic features to describe the nature of these modifications.

4. PHONETIC FEATURES

You have already noticed that sound segments, in systematic phonetics, are
said to be composed of bundles of features. In fact, the identification of sound
segments on preceding features is based on the binary features of those
sounds that we have put within square brackets. You have also noted that the
set of features used to identify vowels differs from the set used to identify other
types of phonemes. Now, it is important to know what we mean by these
features.

4.1. VOCALIC VERSUS CONSONANTAL

There are a good number of phonetic features. However, two of them are
considered to be the most crucial ones: (a) [±consonantal], and (b) [±vocalic].
Their importance is in that they classify all sounds (in the case of English and
Farsi at least) into the four major sound classes: (a) consonants, (b) vowels, (c)
liquids, and (d) glides:

VOWELS CONSONANTS
+vocalic -vocalic
-consonantal +consonantal

LIQUIDS GLIDES
+vocalic -vocalic
+consonantal -consonantal

The distinction between [+consonantal] and [+vocalic] is quite simple. If the air,
once out of the glottis, is allowed to pass freely through the oral cavity, the
sound is a [+vocalic] phoneme. If the air, once out of the glottis, is partially or
totally obstructed in one or more places in the oral cavity, the sound is a

80 CHAPTER FOUR
[+consonantal] phoneme. It should be noted that the line between [+vocalic]
and [+consonantal] phonemes cannot be clearly drawn. A continuum exists
between the two extremes. In English, there are also intermediate instances:
(a) liquids, and (b) glides. Liquids share features of both [+consonantal] and
[+vocalic] sounds. Glides lack features of both [+consonantal] and [+vocalic]
sounds.

In order for a phoneme to be [+vocalic], it should meet certain vowel-hood

criteria. These criteria include: (a) the degree of openness of the oral cavity
also known as the degree of aperture, (b) the degree of tension or laxity of the
vocal tract muscles, and (c) amount of duration or length of articulation. There
is a large degree of freedom in the articulation of open vowels. However, this
freedom is not endless. On the one hand, no vowel can be more open than the
standard open vowels (fourth degree of aperture). On the other hand, a vowel
could not be much more "close" than the standard close vowels (first degree of
aperture). Additionally, the close vowels must have a certain minimum duration
in order to be perceived as [+vocalic] rather than [+consonantal].

4.2. VOICE

A sound is described as [-voice] when the vocal cords do not vibrate during its
articulation. If the vocal cords do vibrate, the sound is considered [+voice]. It
was mentioned earlier that the vocal cords are folds of muscle (membranes)
located at the level of the glottis (in fact, the glottis is nothing other than the
space between the vocal cords). The vocal cords vibrate when they are closed
to obstruct the airflow through the glottis. They vibrate under the pressure of
the air being forced through them by the lungs. The [±voice] opposition is
mainly useful for the classification of consonants. Because all vowels are
[+voice], it is considered a redundant feature for vowels that can be captured
by a redundancy rule. In English, the phonemes /ʧ/, /s/, /p/, /k/, /f/, /ʃ/, /t/, /θ/,
and /h/ are [-voice]. Other phonemes are [+voice].

CHAPTER FOUR 81
4.3. NASAL AND NASALIZATION

The top of the pharynx is like a crossroads. The airstream can exit the pharynx
in either of two ways, depending on the position of the soft palate. If the soft
palate is lowered, a portion of the air will pass through the nasal cavity (the
remainder finding its way through the oral cavity). If the soft palate is raised,
access to the nasal cavity is cut off, and the air can only pass through the oral
cavity. The sounds produced via the first method are [+nasal]. Those produced
the other way are [-nasal].

Nasality primarily concerns consonants. However, vowels can be nasalized in

specific contexts (i.e., when they are followed by nasal consonants). Therefore,
the term nasality is used for consonants and nasalization for vowels.
Nasalization is fully explained in the following chapters.

NASAL CONSONANTS NON-NASAL CONSONANTS

Figure 4.1: Nasal versus non-nasal phonemes

[+nasal] phonemes of the vast majority of the world's languages are [+voice].
So, the value [±] for the voicing feature is predictable through a redundancy
rule when a phoneme is [+nasal]. Therefore, [+voice] is not included in the
bundle of distinctive features of a [+nasal] sound segment.

82 CHAPTER FOUR
During the production of these nasal stops, the soft palate is lowered to a
greater or lesser extent, allowing a portion of the airstream to pass through the
nasal cavity. Occlusion (i.e., air blockage) occurs in the mouth only. Many
linguists, therefore, consider [+nasal] phonemes to be stop sounds. However,
[+nasal] phonemes can endure through the nasal cavity. In other words, the
nasal resonance is continuous. Although a few linguists rank the nasals among
the [+continuant] phonemes, the majority of them categorize [+nasal] sounds
as [-continuant].

/m/ /n/ /ŋ/

The English language has three [+nasal] sounds in its inventory of phonemes:
/m/, /n/, and /ŋ/. Farsi only includes the first two: /m/, and /n/. The phoneme /m/
is a bilabial nasal. The mouth is configured just as for the articulation of the
corresponding [-nasal] bilabial stops /p/, and /b/. The lips are pressed tightly
together, but the velum is somewhat lowered to let pulmonary air escape
through the nasal cavity. The vocal cords are kept close to each other to create
audible vibration. The phoneme /n/ is a dental or alveolar nasal. The mouth is
configured just as for the corresponding dental or alveolar stops /d/ and /t/. The
tongue makes contact either with the front teeth, or with the alveolar ridge
directly above them. The phoneme /ŋ/ is a velar nasal. The configuration of the
mouth is very close to that of the corresponding velar stops /g/ and /k/. The

CHAPTER FOUR 83
tongue tip rests against the lower teeth while the back of the tongue makes
contact with the soft palate. However, as the soft palate is lowered (to allow air
to flow through the nasal cavity), the tongue's back is detached from the soft
palate to let a burst of air flow through the mouth.

4.4. ABRUPT RELEASE

Pulmonic egressive air is modified in the oral cavity to produce the phonemes
of certain languages. In fact, there are three major types of air modifications:
(a) complete blockage followed by sudden release, (b) complete blockage
followed by gradual release, and (c) constriction or narrowing of the passage of
air. The first results in the production of stops and plosives, the second in the
production of affricates, and the last in the production of fricatives.

In systematic phonetics, the feature [±abrupt release] indicates whether the

first of these types of air modifications is present in the articulation of a
phoneme or not. In other words, when a phoneme is articulated by a complete
blockage followed by a sudden release of the air in the oral cavity, that
phoneme is [+abrupt release]. Otherwise, the phoneme is [-abrupt release].
Therefore, all stop consonants (including nasals /m/, /n/, /ŋ/, and plosives /p/,
/b/, /t/, /d/, /k/, /g/) have the feature [+abrupt release].

Although nasal consonants are [+abrupt release], this feature is considered a

redundant feature for them; therefore, it is not part of the inventory of features
that are used to identify nasal consonants. In fact, the feature [+nasal] results
in a [+abrupt release] phoneme. This is, like it was mentioned earlier, usually
captured in a redundancy rule. Remember from preceding sections and
chapters that a redundancy rule is a schematic representation (or a specific
notation) that indicates a specific relationship between two entities.
Redundancy rules usually go by the general format of

X ➙ Y

84 CHAPTER FOUR
where X represents one of the entities and Y the other. In phonetics, for
example, X stands for one phoneme or one phonetic feature, and Y for another
phoneme or phonetic feature. The symbol (is read as "rewrites as." Take the
following example:

[+nasal] ➙ [+abrupt release]

This rule means that if a phoneme contains the feature [+nasal], it

"automatically" contains the feature [+abrupt release]. Therefore, the feature
[+abrupt release] does not need to be specifically mentioned in the bundle of
features used to identify a [+nasal] phoneme. In fact, redundancy rules tell us
how we can predict phoneme qualities and the basis of the qualities we have
already identified.

Also notice that the features [abrupt release] and [continuant] are the opposite
of each other. That is, when a phoneme is [+abrupt release], it will certainly be
[-continuant]. This relationship, too, can be represented in the form of the
following redundancy rule;

[+abrupt release] ➙ [-continuant]

4.5. CONTINUANT

Whenever the pulmonic egressive airstream is constricted in the oral cavity, it

is possible to lengthen the duration of the articulation of phonemes that are
produced in that way. In fact, the phonetic feature [continuant] is used to
describe the degree of obstruction involved in the articulation of a sound. If the
obstruction is partial, with enough space for the air to pass out through the
mouth, the phoneme involved is [+continuant]. Otherwise, the phoneme will be
[-continuant]. In English, all vowels, glides, liquids, and fricative consonants are
[+continuant]. Only affricates, stops, and nasal are [-continuant]. The feature
[+continuant] is a redundant feature for liquids and vowels. This is because the
feature [+vocalic] means that the phoneme is automatically [+continuant].

CHAPTER FOUR 85
Therefore, [+continuant] is not included in the bundle of features used to
identify [+vocalic] phonemes. The relationship between [+vocalic] and
[+continuant] phonemes can be shown in the following redundancy rule:

[+vocalic] ➙ [+continuant]

4.6. ASPIRATION

In the articulation of word-initial voiceless stop consonants, in English, there is

a build up of air pressure in the oral cavity prior to the actual release of
airstream at the place of articulation. Therefore, when these phonemes are
produced, an increased puff of air accompanies the phonemes. This quality is
called aspiration. If you light a match, hold it in front of your mouth, and say a
word like Peter, you may be able to feel the aspiration of the first sound.

In broad transcription (see chapter 2), aspiration is not identified; in narrow

transcription, however, a superscript h is used to identify the aspiration of the
phoneme:

toe [thəʊ] kill [khɪɫ]

Peter [phi:tə] chat [ʧhæt]

As you see, in these examples the feature [+aspiration] is represented with a

small superscribed h. The following phonological rule identifies aspirated
phonemes of English:

-voice
➙ [+aspiration] ⁄ # ____
-continuant

The rule tells us that when [-voice, -continuant] phonemes occur at the
beginning of a word (i.e., occupy the word-initial position), they are always
[+aspiration]. This rule is not a redundancy rule because it identifies a condition
for the change to take place. That is, the "word-initial position" condition must
first be met if the [-voice, -continuant] phoneme wants to become [+aspiration].

86 CHAPTER FOUR
Therefore, unlike redundancy rules, phonological rules need specific
conditions. They are if…then rules. The schematic representation of
phonological rules (also called phonological processes) is:

X ➙ Y ⁄ Z

The rule tells us that X will rewrite as Y if the conditions identified by Z are met
(that is, X rewrites as Y in the context of Z). In the above example, the
phoneme which is [-voice, -continuant] rewrites as [+aspiration] in the context
of word-initial position. Phonological and redundancy rules will be treated with
greater detail in the following chapters.

When phonemes containing [-voice] and [-continuant] features occur at the end
of a word, they are sometimes [+aspiration]. When they occur after /s/, these
phonemes are always [-aspiration]. The absence of aspiration can easily be felt
by comparing the pronunciation of pairs of words like pie/spy, toe/stow, kill/skill,
and so on. Also notice that there is never any puff of air accompanying the
articulation of English phonemes which share such phonetic features as
[+voice, -continuant], or [-voice, +continuant], or [+voice, +continuant]. Such
English phonemes are [-aspiration]. Thus, fricatives, nasals, liquids, glides,
vowels, voiced affricates, and voiced stop consonants are always [-aspiration].

4.7. ANTERIOR

The feature [anterior] is related to the place or position of articulation of

phonemes.

Any phoneme which is articulated at a position closer to the frontal parts of the
mouth is [+anterior]; any phoneme articulated at a position closer to the dorsal
(or the posterior) parts of the mouth is considered to be [-anterior].

As figure 4.2 shows, all bilabial, labio-dental, dental, alveolar, and palatal
sounds are [+anterior]. In contrast, all palato-velar, velar, velo-uvular, uvular,
pharyngeal, and glottal sounds are [-anterior].

CHAPTER FOUR 87
 Figure 4.2: Anterior and posterior positions of articulation

In English, liquids are [+anterior] while glides (also less technically called semi-
vowels) are [-anterior].

+vocalic
➙ [+anterior]
+consonantal

-consonantal
➙ [-anterior]
-vocalic

In addition the phonemes /ʃ/, /Ʒ/, /ʧ/, /ʤ/, and /ŋ/ are [-anterior]. Other English
phonemes are [+anterior]. Some phoneticians may use the feature [posterior]
instead of the feature [anterior]. In other words, they use [-posterior] instead of
[+anterior], and [+posterior] instead of [-anterior].

[±anterior] ➙ [∓posterior]

88 CHAPTER FOUR
These rules can be collapsed together to form a single and more general rule.
If we use the mathematical symbol α to represent ±, then we can rewrite the
above two rules as:

[α anterior] ➙ [-α posterior]

In most cases, however, phoneticians normally prefer not to use the feature
[posterior]. The students' attention is drawn to this point so that confusions will
not arise if the feature [posterior] is in any writing.

4.8. CORONAL

Like the feature [anterior], the feature [coronal], too, is related to the place or
position of articulation of phonemes. The feature [coronal], however, shows
whether the tip and the blade of tongue are involved in the articulation of
phonemes or not. Any phoneme which is articulated by the involvement of the
tip (i.e., apex) or the blade of tongue is considered to be [+coronal]; any
phoneme which is not articulated through the involvement of the tip or the
blade of tongue is considered to be [-coronal].

It can easily be felt that usually three parts of the human tongue are involved in
the production of phonemes: (a) the very tip or apex, (b) the blade, and (c) the
back or dorsum:

All interdental, alveolar, and palatal consonants are [+coronal]. Vowels and
liquids are [-coronal]:

+vocalic
➙ [-coronal]
±consonantal

The coronal phonemes of English are /θ/, /ð/, /t/, /d/, /n/, /s/, /z/, /ʒ/, /ʃ/, /ʧ/, /ʤ/,
/l/, /r/, and /j/. In formal notations, the features [+anterior] and [coronal] usually
occur together; in other words, whenever a phoneme is [+anterior], it is usually
necessary to identify whether it is [+coronal] or [-coronal].

CHAPTER FOUR 89
 Figure 4.3: Different parts of human tongue

4.9. STRIDENT

This feature is in fact related to the manner of articulation of phonemes. As you

will certainly remember, consonants are produced when a certain type of
obstruction is introduced to the passage of air in the mouth. Some consonants
are explosive in the sense that the blocked air is suddenly released; some
others are fricatives or affricates because the air moves out of the mouth
through a narrow opening. In this latter case, sometimes there is a turbulence
in the air (like in the case of /s/, /z/, /ʃ/, /ʒ/, /ʧ/, /ʤ/, /f/, and /v/), and sometimes
there is no turbulence (like in the case of phonemes other than these). The
feature [strident] simply says whether this turbulence exists or not. English
liquids, glides, and vowels are [-strident].

[+vocalic] ➙ [-strident]

-vocalic
➙ [-strident]
-consonantal

90 CHAPTER FOUR
Stridency is a feature of fricative (sometimes called spirant) and affricate
consonants. Therefore, plosive consonants (i.e., those with the [+abrupt
release] feature) are predictably [-strident]:

[+abrupt release] ➙ [-strident]

[+Strident] sounds are categorized into two subtypes based on their place of
articulation. /f/ and /v/ are [-sibilant]. However, /s/, /z/, /ʃ/, /ʒ/, /ʧ/, and /ʤ/ are
[+sibilant]. In fact, the term sibilant is used for hushers and hissers. During the
articulation of [+sibilant] sounds, a groove is formed on the tongue so that the
outgoing air takes on a hissing overtone. This hissing sound is not felt during
the articulation of /f/ and /v/ (which are [+strident] labiodentals).

SIBILANT NONSIBILANT

Figure 4.4: Sibilant versus non-sibilant phonemes

4.10. LATERAL

To produce a lateral sound, air is obstructed by the tongue at a point along the
centre of the mouth but the sides of the tongue are left low so that air can
escape over its sides. In fact, the tongue is strongly flexed and the air is forced
through a narrow oval cavity, producing a hushing sound. /l/ is the clearest
example of a [+lateral] sound in English. Both the clear [l] and the word-final
dark [ɫ] allophones (i.e., the variants of the same phoneme) of /l/ are [+lateral]
sounds. All other sounds of the English language are [-lateral]. Since /l/ is a
liquid, we can safely claim that vowels, consonants, and glides are all [-lateral]:

CHAPTER FOUR 91
 [-vocalic] ➙ [-lateral]

+vocalic
➙ [-lateral]
-consonantal

When an alveolar plosive is followed by the lateral /l/, then what happens is that
we simply lower the sides of the tongue to release the compressed air, rather
than raising and lowering the blade of the tongue. If you say the word 'bottle' to
yourself you can feel the sides of the tongue lower to let out the air. Also notice
that, since /l/ and /r/ are the only liquid sounds (i.e., [+vocalic] and
[+consonantal] sounds) of English, the feature [lateral] is the most important
distinctive feature that serves to distinguish between them.

4.11. SONORANT

The nonnasal stops, the fricatives, and the affricates form a major class of
sounds technically termed obstruents. Nasals, on the other hand, form a class
of sounds called sonorants. When the passage of air is blocked or constricted
in the mouth, and the air passes out through the nose, the sound is
[+sonorant]. When, by way of contrast, the air cannot escape through the nose,
the sound is [-sonorant]. Fricatives are continuant obstruents because (a) the
air is constricted in the mouth, and (b) the air cannot escape through the nose.
Nonnasal stops and affricates are noncontinuant obstruents because (a) the air
is completely blocked in the mouth, and (b) it cannot escape through the nose.
Obstruents are, therefore, [-sonorant]. Notice that the feature [+sonorant] is a
redundant feature for [+nasal] sounds:

[+nasal] ➙ [+sonorant]

Also notice that [+sonorant] phonemes are redundantly [-strident].

[+sonorant] ➙ [-strident]

92 CHAPTER FOUR
4.12. SYLLABIC

Vowels usually constitute the "main core" or the nucleus of syllables; every
vowel is at the center of a single syllable. Vowels can even occur as syllables.
For instance, the word ago /əˈgəʊ/ is composed of two syllables: /ə/ and /gəʊ/.
The first syllable is a single vowel. Even in the second syllable, the vowel (or
the diphthong) takes the prominence (or stress) on itself. Therefore, vowels are
said to be [+syllabic].

It was pointed out earlier that every language of the world contains the two
basic classes of speech sounds often referred to by the cover terms
consonants and vowels. This classification implies that consonants and vowels
differ. In the production of consonants the flow of air is obstructed as it travels
through the mouth. Vowels are produced with no obstruction in the mouth
whatsoever. Oral and nasal stops, fricatives, affricates, liquids, and glides all
have some degree of obstruction and are therefore consonants.

Consonants do not correspond exactly to the sounds specified as

[+consonantal], because glides are [-consonantal], forming a subclass with
vowels (which is why they often occur as part of a vocalic diphthong, and are
also called semi-vowels). However, unlike vowels, glides are produced with
some small oral obstruction and therefore do not constitute syllable peaks (i.e.,
they can neither occur as syllables nor do they govern stress), as do vowels.
Vowels, like glides, are [-consonantal] and [+sonorant]. They differ from glides
because they constitute syllable peaks; so vowels are [+syllabic], whereas
glides are [-syllabic].

Liquids and nasals can also be syllabic—function as a syllable. Liquids and

nasals are not always syllabic; rather, they can be syllabic. That is:

(a) They may constitute separate syllables, as shown by the words Rachel
[reɪʧļ], faker [feɪkŗ], rhythm [rɪðm̩], and button [bʌtņ]. These words could
also be pronounced as /ˈreɪʧəl/, /ˈfeɪkər/, /ˈrɪðəm/, and /ˈbʌtən/.

CHAPTER FOUR 93
 (b) They may be nonsyllabic, as in the words lead /li:d/, read /ri:d/, deal /di:l/,
name /neɪm/, or mean /mi:n/.

Similarly, the vowel sound in words like bird and verb are sometimes written as
a syllabic /r/ (i.e., [bŗd] and [vŗb]); most dictionaries, however, prefer to
transcribe syllabic vowels as /ə/ (e.g., [bərd], [vərb]). This is the only instance
where a schwa represents a stressed vowel. The diacritic symbol [ ˌ ] usually
appears under the phoneme to represent its [+syllabic] feature.

SYMBOL MEANING EXAMPLE

[ļ] syllabic l Rachel [reɪʧļ]

[ŗ] syllabic r faker [feɪkŗ]

[ m̩ ] syllabic m rhythm [rɪðm̩]

[ņ] syllabic n button [bʌtņ]

Table 4.4 shows how four different classes of speech sounds may be
distinguished by the features [± consonantal], [± syllabic], and [±sonorant].

obstruents vowels glides nasals and liquids

consonantal + - - +
syllabic - + - ±
sonorant - + + +
Table 4.4: Specifications of major classes of sounds

By the system shown in Table 4.3, obstruents and vowels are distinct classes;
they do not share any feature. Glides are like consonants in the sense that they
are in the class of [-syllabic] sounds, but they are like vowels in the sense that
they are [-consonantal] and [+sonorant]. Similarly, liquids and nasals are in the
[+consonantal] class with obstruents, but share the feature [+sonorant] with
vowels (and sometimes are [+syllabic]). Nonsyllabic liquids, nasals, and glides
(or semivowels) are in the class of [+sonorant, -syllabic] sounds.

94 CHAPTER FOUR
The syllabicity of liquids and nasal consonants may also be shown by
describing the words in which they function as syllables as having short vowels
before the liquids. So, whenever a short vowel precedes a liquid or a nasal
consonant, the nasal consonant or the liquid becomes [+syllabic].

+vocalic
[+nasal] ➙ [+syllabic] ⁄ -consonantal ___
-length

+vocalic
+consonantal
➙ [+syllabic] ⁄ -consonantal ___
+vocalic
-length

In many English dictionaries, short vowels that occur before syllabic phonemes
are placed within parentheses to indicate that they are optional. Take the
following examples:

medal /med(ə)l/ button /bʌt(ə)n/

In these cases, the schwa can be deleted due to the syllabicity of the
phonemes that follow it. As it was mentioned earlier, this is the only instance
where a schwa represents a stressed vowel.

EXERCISE

1. The following sets of minimal pairs show that English /p/ and /b/ contrast in
initial, medial, and final positions.

Initial Medial Final

pit/bit rapid/rabid cap/cab

Find similar sets of minimal pairs for each pair of consonants given:

A /k/-/g/ D /b/-/v/ G /s/-/ʃ/

B /m/-/n/ E /b/-/m/ H /ʧ/-/ʤ/
C /l/-/r/ F /p/-/f/ I /s/-/z/

CHAPTER FOUR 95
2. For each group of vowels listed, state the phonetic feature(s) they all share.

VOWEL GROUP SHARED FEATURE(S)

A /æ/ /ɑ:/

B /ʌ/ /u:/ /ʊ/ /ɔ:/ /ɒ/ /ɑ:/

C /u:/ /ʊ/ /ɔ:/ /ɒ/

D /i:/ /e/ /u:/ /ɔ:/

E /i:/ /ɪ/ /u:/ /ʊ/

3. For each group of sounds listed, state the phonetic feature(s) they all share.

SOUND GROUP SHARED FEATURE(S)

A /m/ /n/ /ŋ/

B /ʧ/ /s/ /p/ /k/ /f/ /ʃ/ /t/ /θ/ /h/

C /f/ /v/ /θ/ /s/ /r/ /w/

D /h/ /w/ /j/

E /ʤ/ /z/ /b/ /g/ /v/ /ʒ/ /d/ /ð/

F /l/ /r/

4. What phonetic property or feature distinguishes the following pairs of

sounds?

SOUND PAIR FEATURE SOUND PAIR FEATURE

A /m/ /n/ F /ʧ/ /ʤ/

B /j/ /w/ G /ɔ/ /ɒ/

C /p/ /b/ H /θ/ /ð/

D /j/ /h/ I /m/ /ŋ/

E /g/ /d/ J /ŋ/ /n/

96 CHAPTER FOUR
5. List the distinctive features of each of the following sounds in the square
brackets provided.

[ph] [ĩ] [ m̩ ]

[ũ] [ŗ] [ņ]

[ļ] [kh] [ʔ]

CHAPTER FOUR 97
6. In systematic phonetics, a word may be described as a linear array of
bundles of phonetic features. Use the same strategy to describe the words
pen, cat, and leak.
pen

cat

leak

98 CHAPTER FOUR
 CHAPTER FIVE
SUPRASEGMENTALS

1. INTRODUCTION

The term suprasegmental refers to those properties of an utterance which are

not properties of any single segment. In chapter four, you were familiarized with
the binary features that sought to distinguish between the different sound
segments of English. Since these features were considered to be the qualities
of sound segments, they are sometimes called segmentals. It is also vital to
note that speech is not a sketchy haphazard use of random sounds. Rather, it
is a more or less continuous, unified, and organized use of sounds that form
syllables which, in turn, form words, phrases, sentences, and so on. Therefore,
it is readily perceivable that features of speech are not confined to the binary
features of sound segments. In fact, there are certain other features that belong
not to any single sound segment, but to groups of them. These other features
are called suprasegmental features, or suprasegmentals.

2. SUPRASEGMENTALS

Phoneticians talk of five major types of suprasegmental features that exist in

almost all languages of the world. They include (a) stress, (b) tone, (c)
intonation, (d) length, and (e) syllable. Notice that while a language like
Chinese makes full use of tone, a language like English does not employ this
suprasegmental feature. Therefore, your attention is drawn to the fact that the
languages of the world make their own choices of suprasegmentals.

2.1. STRESS
The word stress means “loudness.” Stress is a term that we apply to words in
isolation which have more than one syllable. It refers to the property that

CHAPTER FIVE 99
certain syllables carry which makes them stand out from the rest of the word. It
seems certain that stressed syllables are made with more effort than
unstressed ones; the muscles in the lungs seem to expel air more aggressively
than for unstressed syllables. Before we identify what stress means, try this
short exercise: Say this sentence aloud and count how many seconds it takes.

The beautiful Mountain appeared transfixed in the distance.

How much time did it require? Probably about 5 seconds. Now, try saying this
sentence aloud.

He can come on Sundays as long as he doesn't have to do any

homework in the evening.

How much time did this one require? Probably about 5 seconds. The first
sentence is much shorter than the second sentence. Compare them:

The beautiful Mountain appeared transfixed in the distance

He can come on Sundays as long as he doesn't have to do any

homework in the evening

This simple example makes a very important point about how native speakers
speak and use English. Namely, English is considered a stressed language
while many other languages are considered syllabic. What does that mean? It
means that, in English, speakers give stress to certain words (or more precisely
syllables) while other words are quickly spoken. In other languages, such as
French or Italian, each syllable receives equal importance.

Many speakers of syllabic languages do not understand why English people

quickly speak, or swallow, a number of words in a sentence. In syllabic
languages each syllable has equal importance, and therefore equal time is
needed. English however, spends more time on specific stressed words while
quickly gliding over the other, less important, words. For example, native

100 CHAPTER FIVE

speakers quickly glide over the modal verb "can" and it is hardly pronounced
when it is used in the positive form.

They can come on Friday. (stressed words are underlined)

On the other hand, when they use the negative form "can't" they tend to stress
the fact that it is the negative form by also stressing "can't".

They can't come on Friday.

As you can see from the above example the sentence, "They can't come on
Friday" is longer than "They can come on Friday" because both the modal
"can't" and the verb "come" are stressed. So, you see that stress adds to the
length of the syllable that carries it. We mentioned earlier that only some
syllables in a word carry stress. Any good dictionary provides you with
information regarding the stress-carrying syllables of a given word. Take the
following examples:

confine /kənˈfaɪn/
album /ˈælbəm/
commemoration /kəˌmeməˈreɪʃən/

In these examples, primary stress is marked with a raised vertical line, and
secondary (or medium) stress is marked with lowered vertical line. In British
dictionaries, both marks come at the beginning of the stressed syllable where
as in American dictionaries they come at the end of stressed syllables. They
apply to the entire syllable, not to any single segment within the syllable.

2.1.1. THE PLACE OF STRESS

Notice that English has four levels of stress placement: (a) weak or quiet, (b)
tertiary or regular, (c) secondary or louder, and (d) primary or loudest. In the
last example above, only two levels of stress have been identified: primary and
secondary. As you have certainly noticed, only one syllable has remained intact
(or untouched) between the two stressed syllables [kəˌmeməˈreɪʃən]. In fact,

CHAPTER FIVE 101

this observation has an implication: in English, after you identify the primary
stress, you can identify the other levels of stress but you must be careful to
leave one syllable intact between the levels. However, usually only two levels
of stress are identified because no word in English gets long enough to require
more that two levels of stress. We will return to this point in the following
sections.

2.1.2. STRESS VERSUS PROMINENCE

When a syllable receives stress, it becomes prominent. It would have been

logically possible for every syllable to have exactly the same loudness, pitch,
and so on. In fact, some early attempts at speech synthesizers sounded like
this. However, human languages have ways to make some syllables more
prominent than others. A syllable might be more prominent by differing from the
surrounding syllables in terms of (a) loudness, (b) pitch, or (c) length.

Pitch is an auditory sensation that places sounds on a scale from high to low.
Every syllable has pitch, however, any syllable that is articulated with a
noticeably different pitch will be deemed to carry stress. This can go either way:
if all the syllables are said in a low pitch except one, then that higher pitch
syllable will be deemed to carry the stress of the word. Pitch also plays a
central role in intonation. Length, too, seems to play a role in stress. Generally,
if one syllable has a longer length than the others in the word then it is deemed
to be the one carrying stress. Length is one of the more important determiners
of stress. In addition to length and pitch, loudness can result in prominence. It
seems obvious that if one syllable is articulated louder than the others then it
will have achieved some prominence from the other syllables. This prominence
would then make that syllable the stressed syllable. However, it is very difficult
to make a sound louder without affecting the length, pitch or quality of that
syllable. If you could only change the loudness of a sound then the perceptual
change is not as great as you would expect.

102 CHAPTER FIVE

Prominence is relative to the surrounding syllables, not absolute. A stressed
syllable that is nearly whispered will be quieter than an unstressed syllable that
is shouted. In English, the three ways to make a syllable more prominent are to
make it: (a) louder, (b) longer, or (c) higher pitched. The last one (i.e., change
of pitch) is, in fact, the most usual way of achieving prominence in English.
However, English typically uses all three kinds of prominence simultaneously.
Other languages might use only one or two of them. The cues can also be
used differently in other languages. In Swedish, stressed syllables are usually
lower in pitch (which is, in effect, one of the most noticeable features of a
Swedish accent).

Even in English, stress does not always mean higher pitch. In one of the
intonation contours used to convey surprised disbelief, the most strongly
stressed syllable of the utterance has the lowest pitch. Take the following
example:

You're taking phonetics!

One of the most interesting points about stress in English is that vowels in
unstressed syllables are systematically reduced. English speakers will not try to
control the position of the tongue body during the vowel of an unstressed
syllable. Instead, the tongue body will reach whatever point is convenient in
getting from the preceding consonant to the following consonant. The average
position reached is mid-central schwa /ə/. Failing to reduce unstressed vowels
is one of the major contributors to an accent in non-native speakers of English.
Reducing vowels inappropriately is one of the major contributors to an English
accent in other languages. In general, the differences between stressed and
unstressed syllables are more extreme in English than in most languages.

2.1.3. THE IMPORTANCE OF STRESS

There are many ways in which stress can be considered as an important

quality of any language. However, all these ways can be captured under a

CHAPTER FIVE 103

cover term: contrast. The job of stress is to create contrast. In other words,
stress is the safest way to reduce fuzziness in speech.

2.1.3.1. BOUNDARY MARKING

In normal speech, words and phrases simply do not have little pauses between
them. In fact, a spoken sentence will definitely appear like a very long word to a
person who is not familiar with the language. In such a situation, stress or
prominence can help indicate where the boundaries are. This will make life
easier for the listener.

In this connection, it is interesting to notice that a language like French usually

gives prominence to the syllable at the end of a word or phrase. Persian, too,
appears to follow a more or less similar pattern except for its Yazdi accent.
Many other languages, however, give prominence to the initial syllables of
words (e.g., Icelandic, Hungarian). There seems to be a bias for English
listeners to interpret a stressed syllable as the beginning of a new word. As
such, stress is a good help for listeners.

2.1.3.2. ADDITIONAL CONTRASTS

In many languages, changing which syllable is stressed can change the

meaning of a word. Take the following examples:

verb noun
convert /kənˈvɜ:t/ /ˈkɒnvət/
contrast /kənˈtræst/ /ˈkɒntræst/

Here again, the job of stress is to help the listener determine what the speaker
means. Therefore, stress is important in the process of speech perception.

104 CHAPTER FIVE

Listening is not the only aspect of your language learning which is affected by
stress. Stress is also important in the articulation of speech. Let us take an
example to demonstrate how stress affects speech:

The beautiful Mountain appeared transfixed in the distance. (14 syllables)

He can come on Sundays as long as he doesn't have to do any homework in

the evening. (22 syllables)

Even though the second sentence is approximately 30% longer than the first,
the sentences take the same time to speak. This is because there are 5
stressed words in each sentence. From this example, you can see that you do
not need to worry about pronouncing every word clearly to be understood
(native speakers certainly do not). You should however, concentrate on
pronouncing the stressed words clearly.

Now, you can do some listening comprehension or speak to your native

English speaking friends and listen to how they concentrate on the stressed
words rather than giving importance to each syllable. You will soon find that
you can understand and communicate more because you begin to listen for
(and use in speaking) stressed words. All those words that you thought you did
not understand are really not crucial for understanding the sense or making
yourself understood. You can safely conclude that stressed words are the key
to excellent pronunciation and understanding of English. Stress in English will
help you to improve your understanding and speaking skills.

2.1.4. RULES OF STRESS

Many EFL (i.e., English as a foreign language) students find it very difficult to
memorize the pronunciation of each and every English word. In fact, there is
the misconception among EFL (English as a foreign language) students that
they have no choice other than memorizing the pronunciations of words. This is

CHAPTER FIVE 105

far from being correct. Knowing a few rules may make it much easier for you to
identify the exact pronunciation of a lot of English words.

2.1.4.1. CONTENT AND FUNCTION WORDS

Before we go on with the discussion of these rules, we should distinguish

between two major groups of words: content words and function words.
Content words are distinguished from function words by the fact that they carry
the encyclopedic information contained in the sentence. In other words, content
words have their own independent meanings and can be used independently.
They can refer to things, states, qualities, or actions in the outside world. They
have their own referents. Function words, by way of contrast, do not refer. They
do not have independent encyclopedic meanings and can only be used in
connection to content words. They usually signal grammatical relationships
among content words in the sentence.
Content words include:

• (most) principle verbs (e.g. visit, construct)

• Nouns (e.g. kitchen, Peter)
• Adjectives (e.g. beautiful, interesting)
• Adverbs (e.g. often, carefully)

Function words include:

• Determiners (e.g. the, a, some, a few)

• Auxiliary verbs (e.g. don't, am, can, were)
• Prepositions (e.g. before, next to, opposite)
• Conjunctions (e.g. but, while, as)
• Pronouns (e.g. they, she, us)

2.1.4.2. WORD STRESS PATTERNS

English is not a language that follows precise rules for the placement of stress.
This, of course, does not mean that it is not possible to find some rules for the

106 CHAPTER FIVE

placement of stress in English. The unfortunate point is that English has a very
complex set of procedures that determine stress. It should be noted that nearly
all English speakers agree on where stress should be placed in individual
words. This means that the English system of stress does have some method
to its chaos. One main consideration is that the presence of stress results in
the prominence of one syllable over others in a word. This may lead the reader
to the conclusion that mono-syllabic words (i.e., words with one syllable or only
one vowel) may not need stress inside the sentence since there is no other
syllable in them to require any contrast whatsoever.

2.1.4.2.1. MONOSYLLABIC WORDS

Obviously, these words do not present any problems for you because, when
pronounced in isolation, they receive the primary stress on their only syllable.
Notice that we do not show the stress of monosyllabic words. When part of a
sentence, monosyllabic content words are pronounced with full forms of vowels
whereas monosyllabic function words are normally pronounced with the weak
forms of vowels (i.e., with /ə/ or /ɪ/). The vowels of monosyllabic function words
are pronounced fully in two contexts: (a) when the word appears in a sentence-
final position, and (b) when the word is used emphatically (or for purposes of
contrast). You can feel easy to reduce all unstressed vowels (whether part of a
content word or a function word) into /ə/. This is, after all, what many dialects
(local forms) of English do. Take the following examples:

SENTENCE PRONUNCIATION
Jack is from London /ʤæk əz frəm ˈlɒndən/
Is Jack from New York? /əz ʤæk frəm nju:jɔ:k/
No, he is from London. /nəʊ hɪ əz frəm ˈlɒndən/
Where is Jack from? /wərəz ʤæk frɒm/

Notice that in all of these sentences the word 'from' is pronounced as /frəm/
except for the last sentence where it is pronounced as /frɒm/. This is because

CHAPTER FIVE 107

of its sentence-final position. Also notice that the vowel of the second syllable
in the word 'London' has been reduced into the weak form /ə/ due to the fact
that this syllable is not stressed.

2.1.4.2.2. BI-SYLLABIC WORDS

In two syllable words, the choice of stress placement is still somewhat simple:
either the first or the second syllable will be stressed. Bi-syllabic function words
usually receive stress on their last syllable; there are some exceptions, though,
which must be memorized. Take the following examples:

between /bətˈwi:n/ before /bəˈfɔ:/

below /bəˈləʊ/ above /əˈbɒv/
upon /əˈpɒn/

Bi-syllabic content words are stressed in a different way. The following hints
are helpful for the placement of stress in bi-syllabic content words:

Bi-syllabic verbs are usually stressed according to their second syllable:

1) If the second syllable of the bi-syllabic verb contains a long vowel or

diphthong, then the second syllable is stressed. Take the following
examples:

increase /ɪŋˈkri:s/ encroach /əŋˈkrəʊʧ/

2) If the bi-syllabic verb ends with more than one consonant, then the
second syllable is stressed. Take the following examples:

collapse /kəˈlæps/ condense /kənˈdens/

3) If the final syllable contains a short vowel and one or no final

consonant, then usually, the first syllable will be stressed. Take these
examples:

108 CHAPTER FIVE

 open /ˈəʊpən/ envy /ˈenvɪ/

Two syllable adjectives are stressed in the same manner. That is, the three
rules mentioned for verbs hitherto, also apply to bi-syllabic adjectives. Take the
following examples:

discreet /dɪsˈkri:t/ disguised /disˈgaɪzd/

fishy /ˈfɪʃɪ/ handsome /ˈhændsəm/

For adverbs, always notice that the ending –ly does not affect stress.
Therefore, it is easy to put the –ly aside and identify the stress for the
remaining adjective. The ending '-ly' is usually pronounced as /lɪ/. Take the
following examples:

handsomely /ˈhændsəmlɪ/ manly /ˈmænlɪ/

lazily /ˈleɪzɪlɪ/ discreetly /dɪsˈkri:tlɪ/

Bi-syllabic nouns generally follow a different stress placement pattern.

1) If the second syllable contains a short vowel, then the stress usually
comes on the first syllable. Take these examples:

labrum /ˈleɪbrəm/ chimney /ˈʧɪmnɪ/

2) If the bi-syllabic noun does not go by rule 1, its stress will have to be
placed on the second syllable. Nouns that fall in this category are very
rare. Take this example:

increase /ɪŋˈkri:s/

2.1.4.2.3. MULTI-SYLLABIC WORDS

Deciding which syllable to give stress becomes very complicated when words
include more than two syllables. With poly-syllabic or multi-syllabic words

CHAPTER FIVE 109

(sometimes called heavy words), the rules of stress placement start to become
ever more arbitrary—with more exceptions than can easily be explained away.
However, this does not mean that there is no rule for this purpose.

One general hint to remember is that weak syllables never carry stress.
Therefore, any unstressed syllable will reduce its vowel into a weaker one—like
/ə/ or /ɪ/. Another hint is that multi-syllabic function words happen, for the most
part, to carry stress on their final or ultimate syllable. Take the following
examples:

nevertheless /ˌnev(ə)rðəˈles/ nonetheless /ˌnʌnðəˈles/

inasmuch (as) /ˌɪnəzˈmʌʧ/ (/əz/) although /ɔ:lˈðəʊ/

Despite the large amount of research done on the subject of stress, it is still
one area of which phoneticians have little understanding. They have, however,
developed a few rules that can be safely used to identify the stress patterns of
the majority, though not all, of the English poly-syllabic content words. Very
often the exceptions are words borrowed from other languages (which are
normally stressed according to the stress patterns of the languages from which
they were originally borrowed.

These rules, and a few examples for each, are provided here. Notice that these
rules do not identify the stress patterns of all content words. They are not
exhaustive and all-inclusive.

1) Prefixes and the adverbial suffix -ly (used to make adverbs) usually do
not change the pattern of stress. Take the following examples:

capitulate /kəˈpɪʧəleɪt/ recapitulate /ˌri:kəˈpɪʧəleɪt/

iterate /ˈɪtəreɪt/ reiterate /ri:ˈɪtəreɪt/
sensitive /ˈsensətɪv/ oversensitive /ˌəʊvəˈsensətɪv/
moderate /ˈmɒdərət/ moderately /ˈmɒdərətlɪ/
tangible /ˈtænʤɪbl/ tangibly /ˈtænʤɪblɪ/

110 CHAPTER FIVE

 2) Verbs that end in –ate or –ize receive stress on their antepenultimate
(i.e., last but two, or the third from right) syllables. Notice that the
endings are pronounced as /eɪt/ and /aɪz/ respectively. Take the
following examples:

confiscate /ˈkɒnfɪskeɪt/ demonstrate /ˈdemənstreɪt/

evaporate /ɪˈvæpəreɪt/ eradicate /ɪˈrædɪkeɪt/
vaporize /ˈveɪpəraɪz/ recognize /ˈrekəgnaɪz/
romanticize /rəˈmæntɪsaɪz/ sexualize /ˈsekʃʊəlaɪz/

3) Verbs that end in –ify usually take stress on the syllable prior to the –ify
ending. Take the following examples:

ramify /ˈræməfaɪ/ beautify /ˈbju:təfaɪ/

testify /ˈtestəfaɪ/ bokonovskify /ˌbɒkəˈnɒfskəfaɪ/

4) Other verbs usually take stress on their last syllables, unless when the
last syllable contains a short vowel. Take the following examples:

intervene /ˌɪntəˈvi:n/ intercede /ˌɪntəˈsi:d/

5) Adjectives that end in –ate receive stress on their antepenultimate

syllable (like verbs) but the –ate ending is pronounced as /ət/. Take the
following examples:

moderate /ˈmɒdərət/ elaborate /ɪˈlæbərət/

6) Adjectives that end in –ese usually receive stress on the syllable

containing –ese. Take the following examples:

Japanese /ˌʤæpəˈni:z/ Javanese /ˌʤævəˈni:z/

CHAPTER FIVE 111

 7) Adjectives which end in –ious, -uous, -eous, -ieous, -ic, -ical, -ian, -ible,
–ial, or -ive (except for those ending in -tive) usually receive stress on
the syllable prior to these endings. There are a few exceptions in
connection to the ending –ic (e.g., Arabic, lunatic, and rhetoric). Take
the following examples:

impressive /ɪmˈpresɪv/ comprehensible /ˌkɒmprəˈhensɪbəl/

public /ˈpʌblɪk/ spontaneous /spənˈteɪnɪəs/
biblical /ˈbɪblɪkəl/ presumptuous /prɪˈzʌmpʧʊəs/
grammarian /grəˈmeərɪən/ dubious /ˈdu:bɪəs/
secretarial /ˌsekrəˈteərɪəl/

8) Adjectives that end in –able, -al, and –ous usually take stress on their
antepenultimate syllables (i.e., the third from right). Take these
examples:

corporal /ˈkɔ:pərəl/ admirable /ˈædmərəbl/

scrupulous /ˈskru:pjələs/

9) Nouns that end in –ity, -ety, -al, -ion, -ence, -ance, -acy, and –ian
usually take stress on the syllable prior to these endings. Take the
following examples:

piety /ˈpaɪətɪ/ importance /ɪmˈpɔ:təns/

ability /əˈbɪlɪtɪ/ correspondence /ˌkɒrəsˈpɒndəns/
proposal /prəˈpəʊzəl/ recognition /ˌrekəgˈnɪʃən/
historian /hɪsˈtɔ:rɪən/ democracy /dɪˈmɒkrəsɪ/

10) Other heavy nouns usually receive stress on their antepenultimate

syllables. Take these examples:

photography /fəˈtɒgrəfɪ/ democrat /ˈdeməkræt/

112 CHAPTER FIVE

 sovereignty /ˈsɒvrəntɪ/ decubitus /dɪˈkju:bɪtəs/
parameter /pəˈræmɪtə/ diplomat /ˈdɪpləmæt/

11) The endings –ist, and –ism do not change stress. Take the following
examples:

organ /ˈɔ:gən/ organism /ˈɔ:gənɪzəm/

physics /ˈfɪzɪks/ physicist /ˈfɪzɪsɪst/

12) Nouns ending in –ee usually receive stress on the ending itself; there
are some exceptions though. Often the exceptions are those nouns
that include double consonants prior to the –ee ending (e.g.,
committee, coffee, etc.). Take the following examples:

referee /ˌrefəˈri:/ absentee /ˌæbsənˈti:/

flunkee /flənˈki:/ devotee /ˌdevəˈti:/

13) –ly does not change stress. Therefore, for adverbs ending in –ly, the
easiest way is to ignore the –ly ending and to identify the stressed
syllable of the adjective. –ly in adverbs is usually pronounced as /lɪ/.

2.2. INTONANTION

There are other suprasegmental features, besides stress, that govern

appropriate English pronunciation. You will certainly remember that these other
features include intonation, tone, and length. Due to its importance in native-
like pronunciation, the suprasegmental feature 'intonation' will be explained
here. 'Intonation' means when, why and how a speaker chooses to raise or
lower or sustain the pitch of her or his voice at particular points while speaking.
These choices are as much a part of the grammatical system of the language
as, say, using the correct auxiliary verb, or remembering when and where to
place the person markers (I, you, he/she ...) which tell us who is carrying out an
action; who or what is acted upon.

CHAPTER FIVE 113

2.2.1. WHAT INTONATION DOES

Intonation in American English is the tools for achieving at least five important
aims: (a) expressing new information, (b) showing contrast, (c) expressing
meaning, (d) showing pronunciation, and (e) showing mood or personality.
Language conveys very specific information, such as how to get somewhere or
what someone is doing. It can be also used beyond the exact meaning of the
words to indicate how the speaker feels about what he is saying, or how he
personally feels at that moment. Generally speaking, if English is not your first
language, this is where you start running into difficulty. Even if you pronounce
each word clearly, if your intonation patterns are non-standard, your meaning
will probably not be clear. Also, in terms of comprehension, you will lose a
great deal of information if you are listening for the actual words used.

Each language deals with expressing these emotional ranges and contextual
importances in different ways. Some languages, such as French and other
Romance languages, stress the end of a sentence, and then use word order to
indicate an important change. Other languages, such as Chinese, have a pitch
change that indicates different vocabulary words, and then superimpose further
pitch change to change meaning or emotion. Because English has a fairly
strictly fixed word order, it is not an option to rearrange the words when we
want to make a point about something. Intonation in American English is the
rise and fall of pitch in order to convey a range of meanings, emotions or
situations, within the confines of standard grammar and fixed word order.

2.2.1.1. EXPRESSING NEW INFORMATION

This is the starting point of standard intonation. When we say that we need to
stress the new information, it is logical to think, "Hmmm, this is the first time I
am saying this sentence, so it is all new information. I had better stress every
word." Well, not quite. In Standard English, we consider that the nouns carry
the weight of a sentence, when all else is equal. Although the verb carries
important information, it does not receive the primary stress of a first-time noun.

114 CHAPTER FIVE

 Dogs eat bones.

After the information has been introduced, or is being repeated through the use
of pronouns, the intonation shifts over to the verb. Notice how the intonation
changes when a sentence changes from nouns to pronouns:

Dogs eat bones. They eat them.

2.2.1.2. CONTRAST

Once the intonation of new information is established, you will soon notice that
there is a pattern that breaks that flow. When you want to emphasize one thing
over another, you reflect this contrast with pitch change. Notice how the
intonation indicates contrast:

Bob studies English.

Bob studies English, but he doesn't use it.

If a person consistently stresses "contrast words" as opposed to "new

information words", he can end up sounding permanently argumentative or
bad-tempered:

I said it is good.
He doesn't like it. Where are you going?

Additionally, mixed messages occur when modals or verbs of perception are

stressed—you end up with the opposite meaning!

People should exercise more, but . . .

They would help us, if . . .
It looks like chance, but at that price, it's a knock-off.
He seems like a nice guy, but once you get to know him. . .

CHAPTER FIVE 115

2.2.1.3. MEANING

A good exercise to demonstrate the variety of meaning through intonation

changes is to take a single sentence, try stressing each word in turn, and see
the totally different meanings that come out.

1. I didn't say he stole the money.

2. I didn't say he stole the money.
3. I didn't say he stole the money.
4. I didn't say he stole the money.
5. I didn't say he stole the money.
6. I didn't say he stole the money.
7. I didn't say he stole the money.

Once you are clear on the intonation changes in the seven sentences, you can
add context words to clarify the meaning:

1. I didn't say he stole the money, someone else said it.

2. I didn't say he stole the money, that's not true at all.
3. I didn't say he stole the money, I only suggested the possibility.
4. I didn't say he stole the money, I think someone else took it.
5. I didn't say he stole the money, maybe he just borrowed it.
6. I didn't say he stole the money, but rather some other money.
7. I didn't say he stole the money, he may have taken some jewelry.

2.2.1.4. PRONUNCIATION

In any language, there are areas of overlap, where one category has a great
deal in common with a different category. In this case, intonation and
pronunciation have two areas of overlap. (a) The first is the pronunciation of the
letter T. When a T is at the beginning of a word (such as table, ten, take), it is a
clear sharp /t/ sound. It is also clear in combination with certain other letters,

116 CHAPTER FIVE

(contract, contain, etc.) When T is in the middle of a word, between two vowels,
or in an unstressed position, it turns into a softer /d/ sound.

Betty bought a bit of better butter. /bedɪ bɔ:d ə bɪd ə bedə bʌdə/

It is this intonation/pronunciation shift that accounts for the difference between

photography /fəˈtɒgrəfɪ/ and photograph /ˈfəʊdəgræf/. (b) The second overlap
has to do with the syllable (within a word) that receives the prominence of
stress and pitch. In fact, the same syllable that receives the stress, will also
govern pitch in the sentence if it occurs in the last content word, or if it is the
focus of contrast in the sentence.

2.2.1.5. MOOD AND PERSONALITY

This is an extremely important aspect of intonation, as it goes beyond what you

are trying to say--it dictates how your listener will relate to you as an individual.
In fact, intonation determines if you will be considered charming or rude,
confident or nervous, or informed or unfamiliar.

An extremely important part of intonation is inside a one-syllable word. You

may wonder if intonation exists in a one-syllable word at all. However, we often
put little sounds in a one-syllable word that are not in the written language, but
that convey a great deal of information in terms of who we are. These extra
sounds are explained in the following sections under the heading of liaisons.

When we contrast two similar words, one ending with a voiced consonant (d, z,
g, v, b) and the other with an unvoiced consonant (t, s, k, f, p), you will hear the
difference in the preceding vowel, specifically in the length or duration of that
vowel.

Simply put, words that end in a voiced consonant have a doubled vowel sound.
For example, if you say bit, it is a quick, sharp sound—a single musical note. If

CHAPTER FIVE 117

you say bid, however, the word is stretched out, it has two musical notes, the
first one higher than the second, bi-id.

single double

tense beat bead

lax bit bid

2.2.2. TYPES OF INTONATION

Before we embark on any discussion of the types of intonation, it is vital to note

that the last content word in the sentence is usually the one that carries
sentence stress. It is, therefore, the one that affects the intonation of the
sentence. In fact, in its most technical sense, intonation refers to (a) categories
of pitch 'peaks' and 'valleys' as well as their combinations at each sentence
stress position (i.e., the last content word of the sentence), (b) types of pitch
category concatenation, and (c) pitch of sentence fractions occurring before the
first sentence stress.

All vowels with 'primary' or 'secondary' sentence stress, receive intonation

features, which may be either 'valleys' (i.e., falls) or 'peaks' (i.e., rises). In the
case of peaks, they may contain a unidirectional fall, or rise, again at the end
resulting in either a rise-fall or fall-rise intonation pattern. Peaks are
characterized by a quick rise confined to the vicinity of a sentence-stressed
syllable. Peaks are of three types: early-peaks, medial peaks, and late peaks.
When the rise precedes the onset of syllable stress, and is narrow in temporal
extension, an 'early peak' will form. When the rise extends into the first half of
the stressed syllable, it results in a medial peak. In the late peak, the rise starts
after the stressed vowel onset and continues into the second half of the
stressed syllable or beyond. Valleys, on the other hand, have a continuous rise,
starting before the stressed syllable (early) or inside it (non-early) and

118 CHAPTER FIVE

extending as far as the beginning of the following sentence-stressed syllable. If
there are several unstressed syllables between two sentence stresses a valley
is thus realized as more gradual compared with the much quicker rise for a late
peak. The less distance there is between stressed syllables the more difficult it
becomes to distinguish between a "valley+peak" and a late "peak+peak"
sequence.

Since it receives the most important stress of the sentence, the last content
word of the sentence is considered to be the cradle of intonation. Therefore,
the ups and downs of pitch in connection to this content word are often used to
identify the intonation of a sentence. In general, English uses six different forms
of intonation patterns: (a) falling, (b) rising, (c) falling-rising, (d) rising-falling, (e)
take-off, and (f) level.

2.2.2.1. FALLING INTONATION

Falling intonation is characterized by a clear fall of the pitch after the stressed
syllable of the last content word towards the end of the sentence. In other
words, the pitch will have to fall after the stressed syllable of the last content
word. Falling intonation is used in many situations. Statements, WH-questions,
and confirmatory tag questions (i.e., tag questions in which you expect the
listener to confirm what you say) all use falling intonation.

Statements are those sentences that provide information for the listener. They
may be either positive or negative. They may sometimes be imperative. All of
these forms of statements require falling intonation. For statements, you should
stress the nouns and let the tone fall at the end of the sentence (e.g., Dogs eat
bones). Take the following examples:

Open the door, please.

Ali will go to Tehran tomorrow.
We've just come from the airport.
Her father makes sure she doesn't sell favor to friends.

CHAPTER FIVE 119

WH-questions are those questions that ask for information by having the
question word (e.g., where, when, how, what, which, why, who, etc.) placed at
the front of the sentence (e.g., Where did she go?). You have certainly noticed
that the pitch falls for the questions that begin with WH-words like where, when,
etc. Take the following examples:

Hey! Where are you going?

What should we do in Senggigi?
Which temples should I visit?
Who would eat it?
How does her father ensure she doesn't sell favor to friends?
Why not visit the temples at Lingsar?
Where will we buy mangoes next time?

Tag questions are very small questions (usually in the form of auxiliary verbs
followed by subject) that come at the end of statements (often for politeness
purposes). They can receive either falling or rising intonation depending on the
predictions the speaker can make. When the speaker believes that the listener
will confirm what he says, he will use falling intonation; when, on the other
hand, the speaker is not sure whether the listener will confirm his position, he
may prefer to use rising intonation. Rising intonation is often more polite. Take
the following examples:

Jack will come over to lunch, won't he?

You come from London, don't you?
Let's go to the movies, shall we?

2.2.2.2. RISING INTONATION

Rising intonation is characterized by a clear rise of the pitch after the stressed
syllable of the last content word of the sentence onwards. In other words, the
pitch will have to rise after the stressed syllable of the last content word. Rising

120 CHAPTER FIVE

intonation is used in yes-no questions, repeated questions, question-
statements, and tag questions. English clearly differentiates between two types
of questions: WH-questions and yes-no questions. Yes-no questions are simply
those that require a yes or a no answer (with perhaps a little further
elaboration, depending on the question.) In English, yes-no questions are often
differentiated from question-word questions (also called information questions
or WH-questions) in terms of intonation. Yes-no questions require rising
intonation. Take the following examples:

Have you just come from the airport?

Is there a local bus terminal nearby?
Is it possible to get to Flores from Senggigi?
Do you really want to see a monitor lizard?

Notice that pitch on a regular question goes up (compared with a statement),

but drops back down at the end.

Do dogs eat bones?

A repeated, rhetorical or emotional question goes up, and then up again at the
end.

Do dogs eat bones?!

Rising intonation is also used in tag questions in which the speaker is not sure
if the listener will confirm what he says. Take the following examples:

Jack will come over to lunch, won't he?

You come from London, don't you?
Let's go to the movies, shall we?

Sometimes speakers refrain from using direct questions and, instead, modulate
the intonation of statements to change them into questions. In these cases, too,
rising intonation is required. Take the following examples:

CHAPTER FIVE 121

 Ali will go to Tehran tomorrow?
Her father makes sure she doesn't sell favor to friends?

Questions of this type are more of a rhetorical or emotional nature than pure
yes-no questions. They are often used to express the speakers' disbelief in the
ideas expressed by the statement.

2.2.2.3. RISING-FALLING INTONATION

Rising-falling intonation is characterized by a combination of a rise followed by

a fall. Rising-falling intonation is characterized by a combination of a rise
followed by a fall. In this case, the pitch will go up on one part of the sentence
and then fall down on another part. Rising-falling intonation is often used in
two-part statements, intro-phrases, and listings.

In two-part sentences, the first half of the sentence usually sets up the second
half. That is, the first half requires rising intonation, and the second half will
take on falling intonation. The cradle of intonation in both parts is the last
content word of each part. Take the following example:

Dogs eat bones, but cats eat fish.

Intro-phrases, too, require rising-falling intonation patterns. When you want to

preface your statement, use a rising tone. In the following example, the first
part requires rising intonation because it is an intro phrase. The second part will
then be articulated with falling intonation.

As we all know, dogs eat bones.

Rising-falling intonation is also used in listings. With more than one item in a
list, all but the last one have a rising tone. The last item in the list will, however,
require falling intonation.

Dogs eat bones, kibbles, and meat.

122 CHAPTER FIVE

If there are more than two sentences joined by conjunctions, the same pattern
will be used. That is, all but the last one have a rising tone. The last part will,
however, require falling intonation. Take the following example:

Dogs eat bones, cats eat cheese, and mice eat wheat.

In this example, the parts prior to the conjunction "and" require rising
intonation, and the part following it requires falling intonation.

2.2.2.4. FALLING-RISING INTONATION

Falling-rising intonation is characterized by a combination of a fall followed by a

rise. This intonation pattern is the reverse of the rising-falling pattern. Falling-
rising intonation is often used for soothing and politeness. When small children
feel uneasy, for example, parents use falling-rising intonation to soothe them.
Soothing refers to statements that make pain or discomfort less severe. Take
the following examples:

Don't cry.
I'll take you to the park.

For purposes of politeness, too, falling-rising intonation is a good tool. Take the
following example:

Thank you

If the level of pitch falls on "thank" and then rises over "you", the expression will
be more polite than when it is pronounced with falling intonation.

Falling-rising intonation pattern is sometimes used for pleading. Pleading refers

to statements in which the speaker makes an earnest or urgent entreaty, often
in emotional terms. Take this example:

I plead with you to stay.

CHAPTER FIVE 123

You will have certainly noticed that falling-rising intonation is a good tool for the
expression of emotions. Politeness, soothing, and pleading are all emotionally
charged.

2.2.2.5. TAKE-OFF INTONATION

Take-off intonation is characterized by a take-off pattern after the most

important contrastive stress of the sentence. In other words, the speaker
begins with a regular or level tone and then gradually raises the pitch (similar to
the way airplanes run along the runway before take-off). The point at which the
pitch comes to a full rising pattern depends on the speaker's choice. This is
normally the point the speaker finds most important. Take-off intonation is often
used in statements with negatively-charged emotions. For example, take-off
intonation can freely be used for grumbling. Take the following example:

You shouldn't have given him all that money you silly boy.

Grumbling is a kind of emotional expression, one that is normally negative.

Other forms of negative emotional expressions, too, require take-off intonation.
Cursing, blasphemy, etc. often take on this intonation pattern.

2.2.2.6. LEVEL INTONATION

Level intonation is characterized by no or very little pitch modulation throughout

a sentence. In rising or falling intonation patterns (or their combinations), the
pitch would move from one level to another (i.e., a higher or lower level). In
level intonation, however, the modulations of pitch are not remarkable enough
to move across different pitch levels. That is, the pitch will stay within the same
level throughout the statement. Level intonation has a very strict usage. It is
used in places where frozen style is best. One such situation is in the church.
The priest usually uses this intonation pattern when he reads the holy book to
the church-goers. Sometimes, in courts of law, too, this intonation pattern is

124 CHAPTER FIVE

used. Learners of English do not need to worry about this intonation pattern
since they will almost never need it.

2.2.3. PHRASING VERSUS INTONATION

In addition to the intonation of a statement, there is another aspect of speech

that indicates meaning—phrasing. Have you ever caught just a snippet of a
conversation in your own language, and somehow managed to piece together
what came before or after the part you heard? This has to do with your
"grammar of expectancy" or your natural understanding of phrasing. In a
sentence, phrasing tells you where the speaker is at the moment, where he is
going, and if he is finished or not. Notice that the intonation stays on the nouns.

2.3. TONE

So far this chapter has been mainly concerned with the major suprasegmental
aspects of phonology (i.e., stress and intonation). There are, however, other
linguistically important suprasegmentals like tone, length, and syllable.

The way pitch is used linguistically differs from language to language. In

English, it does not much matter whether you say 'cat' with a high pitch or a low
pitch. It will still mean "cat." In other words, tone does not change within a
single syllable. Therefore, English is a non-tonal language. However, this is not
the case in many languages. For example, if you say 'ba' with a high pitch in
Nupe (a language spoken in Nigeria), it will mean "to be sour;" if you say 'ba'
with a low pitch, it will mean "to count." Languages that use the pitch of
individual syllables to contrast meanings are called tone languages (or tonal
languages). In other words, tone refers to the changes of pitch within a single
syllable that result in changes in meaning. Thai is a language that has
contrasting pitches, or tones. The same string of "segmental" sounds
represented by [na:] will mean different things when spoken with a low pitch, a
mid pitch, a high pitch, a falling pitch from high to low, or a rising pitch from low
to high:

CHAPTER FIVE 125

 SYLLABLE TYPE OF TONE MEANING
na low tone a nickname
na mid tone rice paddy
na high tone young maternal uncle or aunt
na falling tone face
na rising tone thick

In some tone languages the pitch of each tone is "level"; in others, the direction
of the pitch (whether it glides from high to low, or from low to high) is important.
Tones that "glide" are called contour tones; tones that do not are called level or
register tones. In Thai, there are three level or register tones (low, mid, and
high) and two contour tones (rising and falling). Since English is not a tonal
language, this brief explanation of tone seems to suffice here.

2.4. LENGTH

Length, too, seems to play a role in stress. Generally, if one syllable has a
longer length than the others in the word, then it is deemed to be the one
carrying stress. Length is one of the more important determiners of stress.

2.5. SYLLABLE

The notion of syllable has already been discussed in chapter two. It is,
therefore, unnecessary to explain it here again. However, your attention is
drawn to one important point: not all random combinations of phonemes can
result in syllables in a language. Any language imposes some limitation on the
types of phonemes that go together to form syllables. These restrictions are
called sequential constraints because they constrain the sequences of sounds
that go together in a syllable. You will remember, for example, that Persian
does not allow initial consonant clusters in words. The same is true about
Persian syllables: Persian does not allow initial consonant clusters in syllables
either.

126 CHAPTER FIVE

 EXERCISE

1. Use the rules of stress discussed above to transcribe the following words. If
the rules do not apply, write "impossible" in the corresponding cell.

WORD TRANSCRIPTION WORD TRANSCRIPTION

syllabic eradicate

photography although

presumptuous rapid

sporadic devotee

parameter redemption

democracy elaborate

chemist corporal

political historian

referee comedy

survival diplomat

2. Say the following sentences with their correct intonation.

How are you today?

Are Dan and Megan studying?
They are in London.
He is from Peru.
Where does your brother work?
Your cough sounds terrible. Take some cough drops.
John will win the price, will he?
What city was your English school in?

CHAPTER FIVE 127

3. Underline the word in each of the following sentences that normally receives
the sentence stress.

How do you spend your day, Helen?

They are perfect for you.
These jackets are nice.
What kind of music do you like?
Tell me about your brother and sister, Sue.
Wow! and what does your brother do?
He has an exhibition there.
What do you do after class?
What city was your English school in?
Do you want to play sometime?
What a wonderful thing democracy is.

4. The utterance 'I thought it would rain' can have two implications based on
the kind of intonation with which it is said:
(a) 'I thought it would rain' implying that it rained.
(b) 'I thought it would rain' implying that 'I am surprised why it did not rain'.
Try saying the sentence with the appropriate intonation for each of these
implications.

5. What is the difference between phrasing and intonation?

6. How is tone different from intonation?

7. What are the different functions of intonation in English?

8. Choose some utterances to illustrate the functions of intonation in English.

128 CHAPTER FIVE

 CHAPTER SIX
PHONEMICS

1. INTRODUCTION

Phonetics describes the observable and concrete aspects of language (i.e.,

speech or sounds). There are other abstract and mental aspects of language
that are not the subject of phonetics. We plan speech and then speak these
plans. When we speak our planned speech, some changes take place due to
biological and physiological constraints.

Phonetics can only describe the end result of our plans—speech. It does not
describe the changes that occur to our plans; neither does it tell us what goes
on in the human brain during the planning of speech. For instance, we all know
that /ʃænbe/ is the correct form of the word meaning Saturday in Persian, yet
we all pronounce it, unconsciously of course, as /ʃæmbe/. This shows that what
we actually speak differs from what we plan to speak. In other words, speech
planning and speech execution are somewhat different stories. That branch of
phonology that is concerned with the explanation of these differences is called
phonemics. As such, phonetics and phonemics are the two sides of a coin
called phonology. Like phonetics, phonemics, too, is of two types: traditional
and modern (or systematic). In traditional phonemics, abbreviations like the
ones used in broad and narrow transcriptions were used. In modern
phonemics, phonetic features are used.

Phonemics is the study of the system that governs the sounds of a language.
Phonemics can also interact with morphology to bring about what is known as
morphophonemics. We all know, for example, that the plural ending "s" is used
in English to pluralize nouns. We also know that this ending can be pronounced
in three different ways: /z/ in dogs, /s/ in books, and /əz/ or /ɪz/ in busses

CHAPTER SIX 129

(usually called the allomorphs of the plural morpheme). This means that the
same morpheme (i.e., the plural morpheme) has three surface realizations
whereas it has only one mental or abstract realization. That is, in our mental
dictionary (or our lexicon), there is only one entry for the plural morpheme;
however, in actual speech, we pronounce this single morpheme in three
different ways. This is because the surrounding sounds affect the surface
realization of this morpheme in speech. Morphophonemics describes the
mental processes that bring about such changes in the pronunciation of
morphemes.

Phonemics accounts for at least two types of knowledge the speakers of a

language possess: (a) the knowledge of sequential/phonological constraints,
and (b) the knowledge of phonological processes. Phonological/sequential
constraints refer to those aspects of human linguistic knowledge that explain
why certain sound combinations are possible while certain other sound
combinations impossible. Phonological processes explain the rules that
account for the differences that exist between planned speech (i.e., phonemic
representation of speech) and executed speech (i.e., phonetic representation
of speech).

2. SEQUENTIAL/PHONOLOGICAL CONSTRAINTS

All native speakers of a language know, unconsciously of course, that the

phonemes of their language cannot be strung together in any random order to
form words. In fact, the phonological system of their language determines
which phonemes can begin a word, end a word, and follow each other. Very
often, sequential constraints are biologically-driven. That is, there is a biological
explanation for them. Since breathing is the primary function of human vocal
organ, speech is considered to be its subsidiary function. Therefore, any form
of speech that can interfere with breathing is impossible; hence, sequential
constraints.

130 CHAPTER SIX

Native speakers of English, for example, tacitly know that English words can
begin with such sequences as sp, sk, dr, tr, and str but not with such
sequences as bp, pb, lbk, etc. This unconscious knowledge tells them that no
English word will ever begin with a sequence that is phonologically constrained
in their language. New words enter the language but they all conform to the
sequential constraints of the language. Wog is phonologically possible and may
be added to the English language in future as a new word but *wgo is
phonologically impossible and will never enter the English language as a new
word. In fact, it can safely be argued that wog is a potential word (a word to
which no meaning has been assigned yet). Such potential words are
sometimes called accidental gaps.

3. PHONOLOGICAL PROCESSES

In connected speech, the usual aim is for ease of communication rather than
complete accuracy. Therefore, speakers unconsciously draw on a number of
phonological processes to aid that ease of communication. There are a good
number of such processes. However, four of them are more important than the
rest: assimilation, elision, neutralization, and linking/joining. Sometimes
speakers make adjacent/close sounds more like each other (assimilation),
sometimes they leave some sounds out altogether (elision), sometimes, they
neutralize the differences between members of a natural class of sounds
(neutralization), and other times they insert an unrelated sound to ease the
transition from one sound to another (dissimilation or joining/linking).

3.1. ASSIMILATION

Assimilation is when one sound is influenced and changed by a neighboring

sound. Many phoneticians use the term assimilation to refer to cases in which
one sound affects the sound that appears on its left side. These scholars do
not refer to right-ward sound change as assimilation.

CHAPTER SIX 131

Generally speaking, there are three types of assimilation that can be identified:
assimilation of place, assimilation of manner, and/or assimilation of voice.
Assimilation of place can be demonstrated by ratbag (that is, [ˈræpbæg]) or
oatmeal (that is, [ˈəʊpmi:ɫ]) where the /t/ sound is replaced by a /p/ sound. This
is because the alveolar plosive /t/ is simplified into the /p/ sound which is closer
to the bilabial plosive /b/. Assimilation of manner refers to when two different
manners of articulation influence each other to form a different type of manner
of articulation. Examples of this are the pronunciation of [Indian] as [ˈɪnʤən] or
the pronunciation of [’səʊlʤə(r)]. This is because the plosive /d/ combines with
the approximant /j/ to form the affricate /ʤ/. Assimilation of voice is illustrated
by the pronunciation of "have to" with an /f/ sound rather than the /v/ sound
(i.e., [ˈhæftə]) as the voiced fricative is followed by a voiceless consonant.

The reason for assimilation is because (a) the tongue cannot always move
quickly enough to get from one position to another in order to articulate the next
sound, or because (b) the mouth is to busy anticipating the following sound. In
either case, it approximates the sound before moving on to the next segment of
sound. This approximation is technically referred to as assimilation. The
following rules explain the most common types of assimilation that take place in
English.

1. When a /d/ sound occurs before a /g/ sound, it is usually converted into
the /g/ sound: (e.g., sad girl [sæggɜ:l])

+consonantal
+consonantal +consonantal
-vocalic
-vocalic -vocalic
+voice
+voice +voice
➙ ∕ ___ +abrupt
+abrupt release +abrupt release
release
-nasal -nasal
-nasal
+coronal -anterior
-anterior

132 CHAPTER SIX

 2. When /t/ is followed by /m/, it usually changes into a /p/ sound. Take
the following example: not me [nɒpmi:].

+consonantal
+consonantal +consonantal
-vocalic
-vocalic -vocalic
-voice
-voice ➙ ∕ ___ +nasal
+abrupt release
+abrupt release +anterior
+anterior
+coronal -coronal
-coronal

3. When /t/ is followed by /p/, it usually changes into a /p/ sound. Take the
following example: that person [ˈðæppɜ:sən].

+consonantal +consonantal
+consonantal
-vocalic -vocalic
-vocalic
-voice -voice
-voice ➙ ∕ ___
+abrupt release +abrupt release
+abrupt release
+anterior +anterior
+coronal
-coronal -coronal

4. When /t/ is followed by /b/, it usually changes into a /p/ sound. Take the
following two examples: that boy [ˈðæpbɔɪ]; light blue [ˈlaɪpblu:].

+consonantal +consonantal
+consonantal
-vocalic -vocalic
-vocalic
-voice +voice
-voice ➙ ∕ ___
+abrupt release +abrupt release
+abrupt release
+anterior +anterior
+coronal
-coronal -coronal

CHAPTER SIX 133

 5. When /d/ is followed by /b/, it usually changes into a /b/ sound (e.g.,
good boy [ˈgʊbbɔɪ]).

+consonantal +consonantal +consonantal

-vocalic -vocalic -vocalic
+voice +voice +voice
➙ ∕ ___
+abrupt release +abrupt release +abrupt release
-nasal +anterior +anterior
+coronal -coronal -coronal

6. When /d/ is followed by /p/, it usually changes into a /b/ sound (e.g.,
good practice [ˈgʊbpræktɪs]).

+consonantal +consonantal +consonantal

-vocalic -vocalic -vocalic
+voice +voice -voice
➙ ∕ ___
+abrupt release +abrupt release +abrupt release
-nasal +anterior +anterior
+coronal -coronal -coronal

3.2. ELISION

Elision refers to when a sound or syllable is lost or omitted. It particularly

affects: (a) consonant clusters, (b) weakly stressed syllables that are not
especially missed and (c) words that end in an alveolar consonant and that are
immediately followed by a word beginning with a consonant. The sounds that
are elided are those sounds that are so weakly articulated that they no longer
have any auditory significance. Contracted forms of words are caused by
elision.

Some elided syllables are represented in standard punctuation, for example,

/I'm/ should be /I am/. In standard speech, the missing vowel is understood and

134 CHAPTER SIX

so meaning does not suffer from this contraction. Elision is one of the reasons
for the great mismatches found in English between a word's spelling and its
pronunciation. [Wednesday], for example, was originally a contraction of
[Odin's day] while today, Odin is barely discernable as the /d/ is no longer
pronounced. Elision is most commonly used in, but is not exclusive to,
connected speech. The faster the speech, the more likely that sounds and
syllables will be elided. Notice that when a vowel is elided, it is usually a weak
vowel, typically the schwa. When a consonant is elided, it is usually because it
is in an environment with other consonants.

One cause of elision is the loss of a weak vowel after the voiceless plosives /p/,
/t/ and /k/. The word permit is often pronounced as [pˈmit] which is an example
of the schwa sound /ə/ being elided after /p/.

-consonantal
+vocalic +consonantal
-low -vocalic
➙ Ø ∕ _____
-high +abrupt release
-back -voice
-tense

Another cause of elision is when a weak vowel is elided before the syllabic
consonants /l/, /m/, /n/ and sometimes /r/. The pronunciation of the word seven
[ˈsevn] demonstrates the loss of the schwa /ə/ before the /n/ sound.

-consonantal
+vocalic
+consonantal
-low
➙ Ø ∕ _____ -vocalic
-high
+syllabic
-back
-tense

CHAPTER SIX 135

Complex clusters are often elided in order to simplify the saying of the sound.
The word clothes /kləʊðz/ is often elided to the much simpler pronunciation
/kləʊz/.

+consonantal +consonantal
-vocalic -vocalic
+voice +voice
➙ Ø ∕ _____
+continuant +strident
-strident +anterior
+coronal +coronal

/v/ is often elided when it comes before a consonant. The Russian proper name
Pavlov is, therefore, sometimes elided to [ˈpæləf] which is a much simpler
pronunciation.

+consonantal
-vocalic +consonantal
+voice ➙ Ø ∕ _____ -vocalic
+strident
-coronal

Some elisions are just by convention or to speed up or simplify the way we

speak. For example in the phrase horse shoe, the /s/ sound is usually elided to
make the pronunciation of the phrase easier; hence, [ˈhɔ:ʃu:]. In other words,
the /ʃ/ sound affects any /s/ or /z/ sound that comes before it. Take the following
examples:

this shoe [ðɪˈʃu:]

for those shoes [fəðəʊˈʃu:z]

136 CHAPTER SIX

 +consonantal
+consonantal
-vocalic
-vocalic
-voice
-voice
➙ Ø ∕ _____ +strident
+strident
+continuant
+anterior
-anterior
+coronal
+coronal

+consonantal
+consonantal
-vocalic
-vocalic
-voice
+voice
➙ Ø ∕ _____ +strident
+strident
+continuant
+anterior
-anterior
+coronal
+coronal

3.3. LIAISONS (JOINING)

If you have ever listened to people speaking a foreign language that was
unknown to you, you may have noticed that it was impossible to pick out
individual words from the string of sounds that you heard. This is because, in
real connected speech, words are linked to one another. This is an especially
important phenomenon in RP where the phoneme /r/ does not occur in syllable-
final position unless the word with a final 'r' is followed by a word beginning with
a vowel. Furthermore, when two vowels meet over a word boundary, an extra
letter is frequently added in order to help the transition. While /r/ is the most
common linking sound, /w/ and /j/ can also be activated between two vowels.
This kind of liaison in which a sound segment is inserted between two other
sound segments to differentiate between them is often called dissimilation.

CHAPTER SIX 137

Dissimilation is the opposite of assimilation. One example of linking is to re-
activate an /r/ sound:

e.g., the teacher of the school /ðə’ti:ʧərəvðə’sku:l/.

Another example of linking is to insert an /r/ between two vowels:

e.g., the idea of /ðə’aɪdɪərəv/.

Linking is not limited to the /r/ sound it can also be used by /j/ and /w/. for
example, in English the [+tense, -low] vowels [i:, e, u:, ɔ:] are predictably
followed by a glide:

+vocalic
-vocalic
-consonantal
-consonantal
+tense
+high
Ø ➙ ∕ -low _____
∝back
∝back
βround
βround

In American English, words are not pronounced one by one. Usually, the end of
one word attaches to the beginning of the next word. This is also true for
initials, numbers, and spelling. Part of the glue that connects sentences is an
underlying hum or drone that only breaks when you come to a period, and
sometimes not even then. You have this underlying hum in your own language
and it helps a great deal toward making you sound like a native speaker. Once
you have a strong intonation, you need to connect all sounds together so that
each sentence sounds like one long word.

The dime easier

They tell me the dime easier.
They tell me the dime easier to understand.
They tell me that I'm easier to understand.

138 CHAPTER SIX

The last two sentences above should be pronounced exactly the same, no
matter how they are written. It is the sound that is important, not the spelling.
This stringing together of sounds is usually called liaison. There are four main
points where liaisons happen in English: (a) when consonants are followed by
vowels; (a) when consonants are followed by other consonants; (c) when
vowels are followed by other vowels; and (d) when /t/, /d/, /s/, and /z/ are
followed by /j/.

3.3.1. CONSONANT-VOWEL SEQUENCES

Words are connected when a words ends in a consonant sound and the next
word starts with a vowel sound, including the semivowels /w/, and /j/ and the
liquid /r/.

Spelling Pronunciation
My name is Ann. /maɪ neɪ mɪ zæn/
American accent /əˈmer k ˈnæksənt/

You also use liaisons in spelling and numbers.

Spelling Pronunciation
LA /eh leɪ/
909-5068 /naɪ nəʊ naɪn, faɪ vəʊ sɪk seɪt/

3.3.2. CONSONANT-CONSONANT SEQUENCES

Words are connected when a word ends in a consonant sound and the next
word starts with a consonant that is in a similar position of articulation. For
example, if a word ends with an alveolar or palatal sound and the next word
starts with another alveolar or palatal sound, these words are going to naturally
join together. This is the same for almost all places of articulation.

CHAPTER SIX 139

 Spelling Pronunciation
I just didn't get the chance /aɪ ʤʌsdɪdən geðə ʧæns/
I've been late twice. /aɪvbɪn leɪtwaɪs/

3.3.3. VOWEL-VOWEL SEQUENCES

When a word ending in a vowel sound is next to one beginning with a vowel
sound, they are connected with a glide (or sometimes the liquid /r/) between
the two vowels.

Spelling Pronunciation
Go away. /gəʊ(w)əweɪ/
I also need the other one. /aɪ(j)a:lsəʊ ni:d ðə(j)ʌðə wʌn/
The idea of space travel is new. /ðəaɪdɪərəv speɪs trævəlɪz nu:/

A glide is either a slight /j/ sound or a slight /w/ sound. How do you know which
one to use? This will take care of itself: the position your lips are in will dictate
either /j/ or /w/. For example, if a word ends in /əʊ/, your lips are going to be in
the forward position, so a /w/ quite naturally leads into the next vowel sound:
/gəʊ(w)əweɪ/. After a long /ɜ:/ sound, your lips will be pulled back far enough to
create a /j/ glide or liaison: /aɪ(j)a:lsəʊ ni:d ðə(j)ʌðə wʌn/. Do not force this
sound too much, though. It is not a strong pushing sound. When one word
ends with /ə/ and the adjacent one begins with /ə/, the use of /r/ is almost
unassailable.

3.3.4. /T, D, S, Z/+/Y/ SEQUENCES

When the /t, d, s, z/ sounds are followed by a word that starts with /j/, both
sounds are connected. These sounds connect not only with written /j/ sounds,
but they do so as well with the initial unwritten /j/ sound of syllables and words.
They form a combination that changes the pronunciation:

140 CHAPTER SIX

a) Use /ʧ/ instead of a /t/ sound followed by a /j/ sound.

Spelling Pronunciation
What's your name? /wʌʧərneɪm/
Can't you do it? /kænʧə du:ɪt/
Don't you like it? /dəʊnʧə laɪkɪt/
actually /æʧəlɪ/

b) Use /ʤ/ instead of a /d/ sound followed by a /j/ sound.

Spelling Pronunciation
What did you do? /wʌ(dɪ)ʤə du:/
Would you help me? /wʊʤʊ help mi:/
Did you like it? /dɪʤə laɪkɪt/
graduation /græʤʊ(w)eɪʃən/

c) Use /ʃ/ instead of a /s/ sound followed by a /j/ sound.

Spelling Pronunciation
insurance /ɪnʃʊrəns/
sugar /ˈʃu:gə/

d) Use /ʒ/ instead of a /z/ sound followed by a /j/ sound.

Spelling Pronunciation
How's your family? /hɑʊʒə fæmɪlɪ/
Who's your friend? /hu:ʒə frend/
casual /kæʒʊwl/
usual /ju:ʒʊwl/

CHAPTER SIX 141

3.4. NEUTRALIZATION

With some pairs of sounds, it is not always clear which of the two sounds is
being articulated. It is as if it lay between the two distinct sounds. This
intermediate sound is known as neutralization. This process can be found in
both vowels and some consonants. The most common examples of
neutralization are the vowels /u/ and /i/ and the consonants /b/, /d/ and /g/ with
their voiceless equivalents /p/, /t/ and /k/. Also, it is sometimes difficult to hear
the contrast in the sounds /s/ and /z/.

When the article 'the' is followed by a vowel, its final vowel is deemed to be /i/.
The schwa is inappropriate, the long /i:/ is too long and the short one /ɪ/ is too
short and so the in-between /i/ is used:

e.g., the orange [ðiˈɒrɪnʤ]

In addition, after some consonants /i/ is used:

e.g., happy [hæpi], easy [ˈi:zi]

Along the same lines, /i/ is used in an unstressed prefix that is followed by a
vowel:

e.g., react [riˈækt]

It is also used in short unstressed words followed by a word beginning with a

vowel:

e.g., he asked [hiˈæskt]

The same is true about the vowel /u/ which is an in-between vowel differing
from both /ʊ/ and /u:/. The vowel /u/ is used when it is in an unstressed syllable
followed by a vowel, or the vowel like consonants (i.e., glides or semi-vowels).
Take the following examples:

e.g. to ask [tuˈæsk] who would [huˈwʊd]

When the plosives [p, t, k, b, d, g] follow /s/ in syllable-initial position, they

become neutralized. An in-between sound is produced which is really difficult to

142 CHAPTER SIX

place as one or the other. In fact, it is only the convention of regular spelling
that determines how we transcribe them. Take the following examples:

The word spill could be transcribed as [sbɪɫ]

The word still could be transcribed as [sdɪɫ]
The word skill could be transcribed as [sgɪɫ]

3.5. VOICE ONSET TIME (VOT)

The term Voice Onset Time (VOT) refers to the timing of the beginning of vocal
cord vibration in CV sequences (i.e., CV syllables) relative to the timing of the
consonant release. The theory proposes that this timing is critical for accurate
perception of the voiced/voiceless phonological contrast between consonants.
This holds for stop consonants and to some extent for fricative consonants
(that is, when the consonant is a plosive and often when the consonant is a
fricative).

In phonology, consonants are contrasted and differentiated on the [voice]

feature. Thus /b, d, g/ are marked [+voice] and /p, t, k/ are marked [–voice] in
phonology. The abstract nature of phonology implies firm boundaries to the
segment, and a straightforward conversion from abstract to concrete
(phonemic representation to phonetic representation) as phonetics realizes
phonology. In fact, it can be shown that in English and many other languages,
vocal cord vibration does not occur throughout a [+voice] stop, and that it does
not begin simultaneously with the beginning of the vowel following a [–voice]
stop.

VOT may be negative, zero or positive. Zero VOT indicates that vocal cord
vibration has begun simultaneously with the release of the plosive consonant;
negative VOT indicates vibration beginning earlier than the release of the
plosive consonant; and, positive VOT indicates vibration beginning after the
release. Different languages have different methods of phonetic realization of
this phonological feature. Notice that zero VOT in French cues the perception

CHAPTER SIX 143

of a [–voice] stop, whereas in English the same auditory cue indicates a
[+voice] stop.

Experiments with VOT have shown that abstract phonological specifications

like [+voice] or [–voice] do not always have a direct one-to-one phonetic
realization. Phonetic features (in this case that of vocal cord vibration) do not
necessarily correspond directly with phonological features (in this case [voice]),
nor do phonetic features necessarily reflect the abstract segmental divisions of
phonology, but tend to 'blur' very often across where a segmental boundary
might be. Notice also that different languages often realize the same abstract
phonological specifications (i.e., phonemic representations) differently in their
phonetics (i.e., phonetic representations). Here the phonological feature
[+voice] is realized in English by a zero VOT, but in French by a comparatively
long negative VOT, whereas the [–voice] feature was realized in English by a
long positive VOT, but in French by a zero or very short positive VOT.

3.6. ASPIRATION

Closely related to the notion of voice onset time (VOT) is the notion of
aspiration. The term aspiration means "pronunciation accompanied by
breathing out." In the studies of voice onset time (VOT) in traditional phonetics,
the focus of attention was on the sound or the airflow which filled the voice
onset time (usually in the form of a puff of air). This quality was usually referred
to as aspiration. When voiceless stop consonants take the word-initial position,
they are often articulated with this added puff of air. Aspiration can be
represented with the formal notations of systematic phonemics as the following
phonological rule:

+consonantal
-vocalic
➙ +aspiration ∕ # ___
-voice
+abrupt release

144 CHAPTER SIX

The symbol # marks word boundaries. If placed to the left (like in the above
rule) it marks the initial position; when placed to the right, it marks the final
position. As such, the above rule reads as: Voiceless stop consonants become
aspirated in word-initial contexts. Compare the /p/ sound in words like pit and
spit.

3.7. NASALIZATION

Nasalization (sometimes called nasal assimilation) occurs when a vowel is

followed by a nasal consonant like in the words moon, room, etc. In these
cases, the vowel usually takes on a nasal quality as a result of the speakers'
anticipation of the following nasal sound. Nasalization can be expressed in the
following phonological rule:

-consonantal
➙ +nasal ∕ ___ +nasal
+vocalic

A special kind of nasalization occurs when one nasal consonant changes into
another nasal consonant. This can be observed in such words as impolite,
imbalance, etc. The prefix [ɪn] can change into [ɪm] when it is followed by
[+anterior, -coronal] consonants. The prefix [ɪn] can also change into [ɪŋ] when
it is followed by [-anterior, -coronal] consonants as in words like incomplete and
inglorious. In fact, the nasal [n] assumes the same value [±] for [anterior] and
[coronal] as the following consonants:

+anterior
➙
∝anterior ∕
_____ ∝anterior
+coronal +nasal
βcoronal βcoronal

Since this kind of nasalization affects the structure of the morpheme in, it is
often the subject of morphophonemics rather than being the subject of
phonemics.

CHAPTER SIX 145

3.8. NON-CONTIGUOUS ASSIMILATION

Assimilation does not always involve sounds that are adjacent. Sometimes, a
sound can affect another sound which is not adjacent or close to it. In cases
where assimilation involves non-adjacent sounds, the term non-contiguous
assimilation may be used. Two such assimilation processes are vowel harmony
and umlaut. In umlauting, the vowel of a root assimilates to the vowel of a
suffix:

-consonantal -consonantal
+vocalic ➙ -back ∕ ___ …∦… +vocalic
+back -back

In vowel harmony the reverse occurs: the vowel of a suffix assimilates to the
vowel of a root:

-consonantal
➙
∝round ∕
∝round …∦… _____
+vocalic
βback βback

The symbol …∦… indicates that the sounds are not adjacent or contiguous
(i.e., some other sounds come in between).

Umlauting is not a feature of English, but many Germanic languages,

Hungarian, Finnish, Turkish, and some other languages make use of this
phonological process. Since umlauting and vowel harmony affect the structure
of morphemes, they are often the subject of morphophonemics.

EXERCISE

1. The following narrow transcriptions show words and non-words in English.

Put an asterisk (*) to the left of the wrong transcriptions to identify them as
non-words.

146 CHAPTER SIX

 Example: *[bæŋ] [bæ͂ŋ]

[pi:tə] [khɪɬ] [bĩ] [ɬu:s] [ɪŋk]

[mu:n] [ræmph] [leɪ] [fəthɒgrəfɪ] [ʧæp]
[lʊk] [metəļ] [bĩ:n] [nũ:n] [θɪŋg]

2. The following are statements of some predictable aspects of the phonetic

features of English. Convert each statement into a formal phonological or
redundancy rule. Use the fewest number of features necessary to identify
the sounds to which the rules apply. Write your answers in the boxes
provided. The first one is done as an example for you.

Example: Nasalize vowels before nasal sounds.

-consonantal
➙ +nasal ∕ ___ +nasal
+vocalic

A: Liquids are lax and voice.

B: Of the nasal consonants, only /m/ and /n/ can occur in word-initial
position.

C: Delete a consonant before a word beginning with any consonant that is

not a glide.

CHAPTER SIX 147

3. Rewrite the following formal phonological and redundancy rules as
statements that explain those phonological processes.

+vocalic
-vocalic
-consonantal
-consonantal
+tense
+high
Ø ➙ ∕ -low _____
∝back
∝back
βround
βround

YOUR ANSWER:

+consonantal +consonantal +consonantal

-vocalic -vocalic -vocalic
+voice +voice -voice
➙ ∕ ___
+abrupt release +abrupt release +abrupt release
-nasal +anterior +anterior
+coronal -coronal -coronal

YOUR ANSWER:

-consonantal
➙
∝round ∕
∝round …∦… _____
+vocalic
βback βback

YOUR ANSWER:

148 CHAPTER SIX

 CHAPTER SEVEN
ACOUSTIC PHONETICS 1

1. INTRODUCTION

Acoustic phonetics is a technical area of linguistics. Phoneticians depict and

analyze sound waves using machines and computer programs. Acoustic
phonetics can be defined as the study of sound waves made by the human
vocal organs for communication. Speech sounds, like sounds in general, are
transmitted through the air as small, rapid variations in air pressure that spread
in longitudinal waves from the speaker's mouth and can be heard, recorded,
visualized, and measured.

Differences between individual speech sounds are directly reflected as

differences in either one or several or all such parameters as duration, pitch,
loudness and quality of the belonging speech waves. Basically, acoustic
phonetics deals with the study and description of three topics: (a) the acoustical
properties of individual phonemes (i.e., speech sounds), (b) prosody, and (c)
voice quality. It forms not only the immediate link between speech production
(i.e., Articulatory Phonetics) and speech perception (i.e., Auditory Phonetics),
but is also important for applications in the fields of signal processing and
speech technology. As such, acoustic phonetics goes hand in hand with an
almost new branch of phonetics called experimental phonetics. Experimental
phonetics also lends support to articulatory phonetics discussed in the following
chapters.

In some ways experimental phonetics, unlike experimental work in other areas

of linguistics, is like experimental work in the physical sciences. For example,
we can (a) examine in detail the acoustic waveform of the sounds of speech

1
Most of the discussions of this chapter were adopted from Encyclopedia Encarta 2004
© 1993-2003 Microsoft Corporation.

CHAPTER SEVEN 149

(i.e., acoustic phonetics) or (b) can examine the behavior of the musculature
during articulation (i.e., articulatory phonetics). The data collected in such
experiments reflect what actually happens when a speaker speaks, and are
thus called real world data.

As you have already noticed, phonetics is interested in discovering what the

speaker actually does when he speaks, not what he feels or thinks about
speech. Of course, people do have feelings about speech and it is important for
linguistics to understand the mental, as well as the physical aspects of
speaking. It has already been mentioned in chapter one that the study of the
underlying mental processes involved in speaking is treated in linguistics for
the most part under the heading "phonology." Phonetics studies what people
actually do when they speak. Of course, the relatively new area of Cognitive
Phonetics tries to characterize some of the mental aspects of speaking which
are not truly phonological. Cognitive phonetics has made the boundary
between phonology and phonetics somewhat fuzzy.

2. WAVE MOTION

Wave Motion, in physics refers to a mechanism by which energy is conveyed

from one place to another in mechanically propagated waves without the
transference of matter. At any point along the path of transmission a periodic
(sine) displacement, or oscillation, occurs about a neutral position. The
oscillation may be of air molecules, as in the case of sound traveling through
the atmosphere; of water molecules, as in waves occurring on the surface of
the ocean; or of portions of a rope or a wire spring. In each of these cases the
particles of matter oscillate about their own equilibrium position and only the
energy moves continuously in one direction.

Such waves are called mechanical because the energy is transmitted through a
material medium, without a mass movement of the medium itself. The only
form of wave motion that requires no material medium for transmission is the

150 CHAPTER SEVEN

electromagnetic wave; in this case the displacement is of electric and magnetic
fields of force in space.

3. TYPES OF WAVES

Waves are divided into types according to the direction of the displacements in
relation to the direction of the motion of the wave itself. If the vibration is
parallel to the direction of motion, the wave is known as a longitudinal wave.
The longitudinal wave (or P wave) is always mechanical because it results from
successive compressions (state of maximum density and pressure) and
rarefactions (state of minimum density and pressure) of the medium. Sound
waves typify this form of wave motion. The successive compressions and
rarefactions (i.e., oscillation) of the particles of matter (or the medium) help the
longitudinal wave to travel. When a person speaks, the molecules of air (or the
medium) oscillate to produce a longitudinal wave that can reach listeners' ears
(see figure 7.1).

Figure 7.1: Longitudinal wave (Reprinted from Encyclopedia Encarta 2004 © 1993-
2003 Microsoft Corporation).

Another type of wave is the transverse (or S wave) wave, in which the
vibrations are at right angles (90 degrees) to the direction of motion. A
transverse wave may be mechanical, such as the wave projected in a taut
string that is subjected to a transverse vibration; or it may be electromagnetic,
such as light, X ray, or radio waves. Some mechanical wave motions, such as

CHAPTER SEVEN 151

waves on the surface of a liquid, are combinations of both longitudinal and
transverse motions, resulting in the circular motion of liquid particles (see figure
7.2).

Figure 7.2: Transverse wave (Reprinted from Encyclopedia Encarta 2004 © 1993-
2003 Microsoft Corporation).

For a transverse wave, the wavelength is the distance between two successive
crests or troughs. For longitudinal waves, it is the distance from compression to
compression or rarefaction to rarefaction. The frequency of the wave is the
number of vibrations per second. The velocity of the wave, which is the speed
at which it advances, is equal to the wavelength times the frequency. The
maximum displacement involved in the vibration is called the amplitude of the
wave. The sequence of a trough and a crest is called a cycle. Frequency is the
technical term used to refer to the number of cycles that occur in a given period
of time (usually one second). Hz (hertz) is the measurement unit that is used to
measure wave frequencies. One Hz is equal to one cycle per second.

4. TYPES OF WAVE MOTION

Waves, such as water or sound waves, are periodic (sine or regular)

disturbances of the medium through which they travel. For longitudinal waves,
the medium is displaced in the direction of travel. For example, air is
compressed and expanded (figure 7.1) in the same direction that a sound wave
travels. For transverse waves, the medium is displaced perpendicular to the
direction of travel. Ripples on the surface of a pond are an example of a
transverse wave: the water is displaced vertically (figure 7.2), while the wave
itself travels horizontally. Earthquakes generate both P (compression, or

152 CHAPTER SEVEN

longitudinal) and S (shear, or transverse) waves, which travel at different
speeds and follow different paths.

5. SOUND

Sound is a physical phenomenon that stimulates the sense of hearing. In

humans, hearing takes place whenever vibrations of frequencies from 15 hertz
to about 20,000 hertz reach the inner ear. The hertz (Hz) is a unit of frequency
equaling one vibration or cycle per second. Such vibrations reach the inner ear
when they are transmitted through air. The speed of sound varies, but at sea
level it travels through cool, dry air at about 1,190 km/h.

The term sound is sometimes restricted to such airborne vibrational waves.

Modern physicists, however, usually extend the term to include similar
vibrations in other gaseous, liquid, or solid media. Physicists also include
vibrations of any frequency in any media, not just those that would be audible
to humans. Sounds of frequencies above the range of normal human hearing,
higher than about 20,000 Hz, are called ultrasonic. Sounds of frequencies
below the range of normal human hearing, lower than 15 Hz, are called
infrasonic.

In general waves can be propagated, or transmitted, transversely or

longitudinally. In both cases, only the energy of wave motion is propagated
through the medium; no portion of the medium itself actually moves very far. In
transverse waves, the material through which the wave is transmitted vibrates
perpendicular to the wave’s forward movement. As a simple example, a rope
may be tied securely to a post at one end, and the other end pulled almost taut
and then shaken once. A wave will travel down the rope to the post, and at that
point it will be reflected and returned to the hand. No part of the rope actually
moves longitudinally toward the post, but each successive portion of the rope
moves transversely. This type of wave motion is called a transverse wave (see
figure 7.3).

CHAPTER SEVEN 153

Figure 7.3: Movement of a transverse wave (Reprinted from Encyclopedia Encarta
2004 © 1993-2003 Microsoft Corporation).

Similarly, if a rock is thrown into a pool of water, a series of transverse waves

moves out from the point of impact. A cork floating near the point of impact will
bob up and down, that is, move transversely with respect to the direction of
wave motion, but will show little if any outward or longitudinal motion.

A sound wave, on the other hand, is a longitudinal wave. As the energy of

wave motion is propagated outward from the center of disturbance, the
individual air molecules that carry the sound move back and forth, parallel to
the direction of wave motion. Thus, a sound wave is a series of alternate
increases and decreases of air pressure. Each individual molecule passes the
energy on to neighboring molecules, but after the sound wave has passed,
each molecule remains in about the same location.

6. PHYSICAL CHARACTERISTICS

Any simple sound, such as a musical note, may be completely described by

specifying three perceptual characteristics: pitch, loudness (or intensity), and
quality (or timbre). These characteristics correspond exactly to three physical
characteristics: frequency, amplitude, and harmonic constitution, or waveform,

154 CHAPTER SEVEN

respectively. Noise is a complex sound, a mixture of many different frequencies
or notes not harmonically related.

6.1. FREQUENCY

Sounds can be produced at a desired frequency by different methods. Sirens

emit sound by means of an air blast interrupted by a toothed wheel with 44
teeth. The wheel rotates at 10 revolutions per second to produce 440
interruptions in the air stream every second. Similarly, hitting the A above
middle C on a piano causes a string to vibrate at 440 Hz (i.e., fundamental
frequency). The sound of the speaker and that of the piano string at the same
frequency are different in quality, but correspond closely in pitch. The next
higher A on the piano, the note one octave above, has a frequency of 880 Hz,
exactly twice as high. Similarly, the notes one and two octaves below have
frequencies of 220 and 110 Hz, respectively. Thus, by definition, an octave is
the interval between any two notes whose frequencies are in a two-to-one ratio
(see figure 7.4).

Figure 7.4: Sound waves with different octaves (Reprinted from Encyclopedia
Encarta 2004 © 1993-2003 Microsoft Corporation).

We perceive frequency as “higher” or “lower” sounds. In the figure above, the

frequency of each higher wave is double that of the one below, producing the

CHAPTER SEVEN 155

same note at different frequencies, from 110.00 Hz to 880.00 Hz. Waves
propagate at both higher and lower frequencies, but humans are unable to hear
them outside of a relatively narrow range (between 15 Hz and 20,000 Hz).

Frequencies of oscillating objects can cover a wide range of values. The

tremors of earthquakes may have a frequency of less than one, while the rapid
electromagnetic oscillations of gamma rays may have frequencies of 1020 or
more. In almost all forms of mechanical vibration, a relationship exists between
frequency and the physical dimensions of the vibrating object. Thus, for
example, the time required by a pendulum to make one complete swing is
partly determined by the length of the pendulum; the frequency or speed of
vibration of a string of a musical instrument is partly determined by the length of
the string. In each instance, the shorter the object, the higher the frequency of
vibration.

In wave motion of all kinds, the frequency of the wave is usually given in terms
of the number of wave crests that pass a given point in a second. The velocity
of the wave and its frequency and wavelength are interrelated. The wavelength
(the distance between successive wave crests) is inversely proportional to
frequency and directly proportional to velocity.

Frequency is expressed in hertz (Hz); a frequency of one Hz means that there

is one cycle or oscillation per second. The unit is named in honor of the
German physicist Heinrich Rudolf Hertz, who first demonstrated the nature of
electromagnetic wave propagation. Kilohertz (kHz), or thousands of cycles per
second, megahertz (MHz), or millions of cycles per second, and gigahertz
(GHz), or billions of cycles per second, are employed in describing certain high-
frequency phenomena, such as radio waves. Radio waves and other types of
electromagnetic radiation may be characterized either by their wavelengths, or
by their frequencies. Electromagnetic waves of extremely high frequencies,
such as light and X rays, are usually described in terms of their wavelength
measure, which is often expressed in angstrom units (symbolized as Å

156 CHAPTER SEVEN

meaning hundred-millionths of a cm). An electromagnetic wave that has a
wavelength of one Å has a frequency of about three billion GHz.

6.2. AMPLITUDE

The amplitude of a sound wave is the degree of motion of air molecules within
the wave, which corresponds to the changes in air pressure that accompany
the wave. The greater the amplitude of the wave, the harder the molecules
strike the eardrum and the louder the sound that is perceived. The amplitude of
a sound wave can be expressed in terms of absolute units by measuring the
actual distance of displacement of the air molecules, the changes in pressure
as the wave passes, or the energy contained in the wave.

Figure 7.5: Sound waves with different amplitudes but same frequency
(Reprinted from Encyclopedia Encarta 2004 © 1993-2003 Microsoft
Corporation).

Ordinary speech, for example, produces sound energy at the rate of about one
hundred-thousandth of a watt. All of these measurements are extremely difficult
to make, however, and the intensity of sounds is generally expressed by
comparing them to a standard sound, measured in decibels. Amplitude is the
characteristic of sound waves that we perceive as volume. The maximum

CHAPTER SEVEN 157

distance a wave travels from the normal, or zero, position is the amplitude; this
distance corresponds to the degree of motion in the air molecules of a wave.
As the degree of motion in the molecules is increased, they strike the ear drum
with progressively greater force. This causes the ear to perceive a louder
sound. A comparison of samples at low, medium, and high amplitudes
demonstrates the change in sound caused by altering amplitude. The three
waves in figure 7.5 have the same frequency, and so should sound the same
except for a perceptible volume difference.

6.3. INTENSITY

The distance at which a sound can be heard depends on its intensity. Intensity
is the average rate of flow of energy per unit area perpendicular to the direction
of propagation, similar to the rate at which a river flows through a gate in a
dam. In the actual propagation of sound through the atmosphere, changes in
the physical properties of the air, such as temperature, pressure, and humidity,
produce damping and scattering of the directed sound waves. You already
remember from chapter two that loudness or intensity is measured in decibels.
The softest audible sound to humans is 0 decibels, while painful sounds are
those that rise above 120 decibels. 0 dB marks the threshold of hearing in
human beings. 120 dB marks the threshold of pain in human beings.

6.4. QUALITY

If A above middle C is played on a violin, a piano, and a tuning fork, all at the
same volume, the tones are identical in frequency and amplitude, but different
in quality. Of these three sources, the simplest tone is produced by the tuning
fork; the sound in this case consists almost entirely of vibrations having
frequencies of 440 Hz. Because of the acoustical properties of the ear and the
resonance properties of the ear's vibrating membrane (ear drum or tympanic
membrane), however, it is doubtful that a pure tone reaches the inner hearing
mechanism in an unmodified form. The principal component of the note

158 CHAPTER SEVEN

produced by the piano or violin also has a frequency of 440 Hz, but these notes
also contain components with frequencies that are exact multiples of 440,
called overtones, at 880, 1320, and 1760 Hz, for example. The exact intensity
of these other components, which are called harmonics (i.e., formant
frequencies or formants), determines the quality, or timbre, of the note. The
technical term "harmonics" refers to a series of subsidiary vibrations that
accompany a primary, or fundamental (i.e., wave-motion) vibration.

Harmonics are most notably observed in musical instruments. They result

when the vibrating body, for example, a stretched string or an enclosed air
column, vibrates simultaneously as a whole and in equal parts (halves, thirds,
fourths, and so on), producing wave frequencies that are in simple ratios with
the fundamental frequency (2:1, 3:1, 4:1, and so on). In musical sound the full-
length vibration produces the fundamental tone (or first harmonic or first
partial), which is usually perceived as the basic pitch of the musical sound. The
subsidiary vibrations produce faint overtones (second and higher harmonics or
partials). As the series progresses, the vibrating segments become smaller, the
frequencies higher, and the musical pitches closer together. The harmonic
series for the tone C is given in the accompanying notation; black notes show
pitches that do not correspond exactly with the Western tuning system.
Harmonics contribute to the ear's perception of the quality, or timbre, of a
sound: On a flute, certain harmonics of the series are most prominent; on a
clarinet, others.

In “overblowing” a wind instrument, the player isolates and makes predominant

one of the higher harmonics, thus extending the range of the instrument

CHAPTER SEVEN 159

upward. Unvalved brass instruments, such as the bugle, produce only the
tones of the harmonic series; the valves of the trumpet and the slide of a
trombone add extra tubing, giving the player a new fundamental wavelength
with a new harmonic series; the instrument can thus produce more tones. The
harmonics of string players are flutelike tones produced when they cause the
string to vibrate solely in halves (or thirds, and so on). Harmonics are a
specialized class of overtones; vibrating membranes and solids, such as bells,
produce complex partials that lie outside the harmonic series.

7. SPEED OF SOUND

The frequency of a sound wave is a measure of the number of waves passing

a given point in one second. The distance between two successive crests of
the wave is called the wavelength. The product of the wavelength and the
frequency equals the speed of the wave. The speed is the same for sounds of
all frequencies and wavelengths (assuming the sound is propagated through
the same medium at the same temperature). The speed of sound in dry, sea
level air at a temperature of 0°C (32°F) is 332 m/sec (1,088 ft/sec). The speed
of sound in air varies under different conditions. If the temperature is increased,
for example, the speed of sound increases; thus, at 20°C (68°F), the speed of
sound is 344 m/sec (1,129 ft/sec). The speed of sound is different in other
gases of greater or lesser density than air. The molecules of some gases, such
as carbon dioxide, are heavier and move less readily than molecules of air.
Sound progresses through such gases more slowly. Conversely, sound travels
through helium and hydrogen faster than through air because atoms of helium
and hydrogen are lighter than molecules of air.

Sound generally moves much faster in liquids and solids than in gases. In both
liquids and solids, density has the same effect as in gases. The speed of sound
also varies directly as the square root of the elasticity of the medium. Elasticity
is the ability of a substance to regain its original shape and size after being
deformed. The speed of sound in water, for example, is slightly less than 1,525

160 CHAPTER SEVEN

m/sec (5,000 ft/sec) at ordinary temperatures—almost five times as fast as in
air. The speed of sound in copper, which is more elastic than water, is about
3,350 m/sec (about 11,000 ft/sec) at ordinary temperatures. In steel which is
even more elastic sound moves at a speed of about 4,880 m/sec (about 16,000
ft/sec). Sound is propagated very efficiently in steel.

8. REFRACTION, REFLECTION, AND INTERFERENCE

Sound moves forward in a straight line when traveling through a medium

having uniform density. Like light, however, sound is subject to refraction,
which bends sound waves from their original path. In polar regions, where air
close to the ground is colder than air that is somewhat higher, a rising sound
wave entering the warmer less dense region, in which sound moves with
greater speed, is bent downward by refraction. The excellent reception of
sound downwind and the poor reception upwind are also due to refraction. The
velocity of wind is generally greater at an altitude of many meters than near the
ground; a rising sound wave moving downwind is bent back toward the ground,
whereas a similar sound wave moving upwind is bent upward over the head of
the listener.

Sound is also governed by reflection, obeying the fundamental law that the
angle of incidence equals the angle of reflection. An echo is the result of
reflection of sound. Sonar depends on the reflection of sounds propagated in
water. A megaphone is a funnel-like tube that forms a beam of sound waves by
reflecting some of the diverging rays from the sides of the tube. A similar tube
can gather sound waves if the large end is pointed at the source of the sound;
an ear trumpet is such a device.

Sound is also subject to diffraction and interference. If sound from a single

source reaches a listener by two different paths—one direct and the other
reflected—the two sounds may reinforce one another; but if they are out of
phase they may interfere, so that the resultant sound is actually less intense
than the direct sound without reflection. Interference paths are different for

CHAPTER SEVEN 161

sounds of different frequencies, so that interference produces distortion in
complex sounds. Two sounds of different frequencies may combine to produce
a third sound, the frequency of which is equal to the sum or difference of the
original two frequencies.

9. TYPES OF ORDINARY SOUND

In speech, music, and noise, pure tones are seldom heard. A musical note
contains, in addition to a fundamental frequency, higher tones that are
harmonics of the fundamental frequency. Speech contains a complex mixture
of sounds, some (but not all) of which are in harmonic relation to one another.
Noise consists of a mixture of many different frequencies within a certain
range; it is thus comparable to white light, which consists of a mixture of light of
all different colors. Different noises are distinguished by different distributions of
energy in the various frequency ranges.

When a musical tone containing some harmonics of a fundamental tone, but

missing other harmonics or the fundamental itself, is transmitted to the ear, the
ear forms various beats in the form of sum and difference frequencies, thus
producing the missing harmonics or the fundamental not present in the original
sound. These notes are also harmonics of the original fundamental note. This
incorrect response of the ear may be valuable. Sound-reproducing equipment
without a large speaker, for example, cannot generally produce sounds of pitch
lower than two octaves below middle C; nonetheless, a human ear listening to
such equipment can resupply the fundamental note by resolving beat
frequencies from its harmonics. Another imperfection of the ear in the presence
of ordinary sounds is the inability to hear high-frequency notes when low-
frequency sound of considerable intensity is present. This phenomenon is
called masking.

In general, reproductions of speech and musical themes are recognizable even

if only a portion of the frequencies contained in the originals are copied.
Frequencies between 250 and 3,000 Hz, the frequency range of ordinary

162 CHAPTER SEVEN

telephones, are normally sufficient. A few speech sounds, such as /ð/ require
frequencies as high as 6,000 Hz. For high quality reproduction, however, the
range of about 100 to 10,000 Hz is necessary. Sounds produced by a few
musical instruments can be accurately reproduced only by adding even lower
frequencies, and a few noises can be reproduced only at somewhat higher
frequencies.

10. SPEECH SIGNAL PROCESSING

Closely related to acoustic phonetics is the notion of speech signal processing.

Analog and digital processing of speech signals belongs to the realm of the
engineering sciences and aims at the extraction of information from the
acoustical speech wave. In addition to developing (and continuously refining!)
methods of direct measurement and observation, researchers in the field of
speech signal processing also deal with the important tasks of signal
representation and signal transformation, both of which are indispensable
prerogatives for a variety of applications such as speech storage and
transmission, speech synthesis, speech recognition, speaker identification and
verification, speech signal enhancement and the development of aids to the
handicapped.

EXERCISE

1. Ensure that you understand the following terms:

waveform amplitude sound quality

cognitive phonetics basic pitch transverse wave
longitudinal wave subsidiary vibrations interference
rarefaction formant noise
cycle intensity harmonic
faint overtone frequency

CHAPTER SEVEN 163

2. How is frequency measured?

3. Define the following terms:

crest
wave length
trough
duration
pitch
loudness

4. The following terms are related to acoustic phonetics, but they do not appear
in this chapter. Search the library or the Internet for books or papers on
acoustic phonetics, and try to find brief definitions for these terms:

periodic vibration
sinusoidal vibration
white noise
damped vibration
resonant frequency
spectral envelop
locus
mel
discrete Fourier transform
phantom fundamental
digitization
quantization
sampling rate

164 CHAPTER SEVEN

 CHAPTER EIGHT
AUDITORY PHONETICS 1

1. INTRODUCTION
In chapter one, phonetics was defined to be a branch of linguistics which is
concerned with the production, physical nature, and perception of speech
sounds. We also divided phonetics into three main branches: auditory,
acoustic, and articulatory. Auditory phonetics is the field involved in determining
how speech sounds are perceived by the human ear.

Ear is the organ of hearing and balance. Only vertebrates, or animals with
backbones, have ears. Invertebrate animals, such as jellyfish and insects, lack
ears, but have other structures or organs that serve similar functions. The most
complex and highly developed ears are those of mammals, or animals that
have breasts.

2. STRUCTURE OF THE HUMAN EAR

Like the ears of other mammals, the human ear consists of three sections: the
outer, middle, and inner ear. The outer and middle ears function only for
hearing, while the inner ear also serves the functions of balance and
orientation.

The outer ear includes the auricle (pinna), the visible part of the ear that is
attached to the side of the head, and the waxy, dirt-trapping auditory canal. The
tympanic membrane (eardrum) separates the external ear from the middle ear
which is an air-filled cavity. Bridging this cavity are three small bones—the

1
Most of the discussions of this chapter were reprinted from Encyclopedia Encarta
2004 © 1993-2003 Microsoft Corporation.

CHAPTER EIGHT 165

malleus (hammer), the incus (anvil), and the stapes (stirrup). The cochlea and
semicircular canals make up the inner ear.

Figure 8.1: Structure of the human ear (Reprinted from Encyclopedia Encarta 2004
© 1993-2003 Microsoft Corporation).

2.1. OUTER EAR

The outer ear is made up of the auricle, or pinna, and the outer auditory canal.
The auricle is the curved part of the ear attached to the side of the head by
small ligaments and muscles. It consists largely of elastic cartilage, and its
shape helps collect sound waves from the air. The earlobe, or lobule, which
hangs from the lower part of the auricle, contains mostly fatty tissue.

The outer auditory canal, which measures about 3 cm (about 1.25 in) in length,
is a tubular passageway lined with delicate hairs and small glands that produce
a wax-like secretion called cerumen. The canal leads from the auricle to a thin
taut membrane called the eardrum or tympanic membrane, which is nearly
round in shape and about 10 mm (0.4 in) wide. It is the vibration of the eardrum

166 CHAPTER EIGHT

that sends sound waves deeper into the ear, where they can be processed by
complex organs and prepared for transmission to the brain. The cerumen in the
outer auditory canal traps and retains dust and dirt that might otherwise end up
on the eardrum, impairing its ability to vibrate.

The inner two-thirds of the outer auditory canal is housed by the temporal
bone, which also surrounds the middle and inner ear. The temporal bone—the
hardest in the body—protects these fragile areas of the ear.

2.2. MIDDLE EAR

The eardrum separates the outer ear from the middle ear. A narrow
passageway called the eustachian tube connects the middle ear to the throat
and the back of the nose. The eustachian tube helps keep the eardrum intact
by equalizing the pressure between the middle and outer ear. For example, if a
person travels from sea level to a mountaintop, where air pressure is lower, the
eardrums may cause pain because the air pressure in the middle ear becomes
greater than the air pressure in the outer ear. When the person yawns or
swallows, the eustachian tube opens, and some of the air in the middle ear
passes into the throat, adjusting the pressure in the middle ear to match the
pressure in the outer ear. This equalizing of pressure on both sides of the
eardrum prevents it from rupturing.

The middle ear is a narrow, air-filled chamber that extends vertically for about
15 mm (about 0.6 in) and for nearly the same distance horizontally. Inside this
chamber is a linked chain of three ossicles, or very small bones. Both the Latin
and common names of these bones are derived from their shapes. They are
called the malleus, or hammer; the incus, or anvil; and the stapes, or stirrup,
which is the tiniest bone in the body, being smaller than a grain of rice.

The hammer is partly embedded in the eardrum, and the stirrup fits into the
oval window, a membrane that fronts the inner ear. Vibrations of the eardrum
move the hammer. The motion of the hammer moves the anvil, which in turn

CHAPTER EIGHT 167

moves the stirrup. As sound vibrations pass from the relatively large area of the
eardrum through the chain of bones, which have a smaller area, their force is
concentrated. This concentration amplifies, or increases, the sound just before
it passes through the oval window and into the inner ear. When loud noises
produce violent vibrations, two small muscles, called the tensor tympani and
the stapedius, contract and limit the movement of the ossicles, thus protecting
the middle and inner ear from damage.

2.3. INNER EAR

The chain of bones in the middle ear leads into the convoluted structures of the
inner ear, or labyrinth, which contains organs of both hearing and balance. The
three main structures of the inner ear are the cochlea, the vestibule, and the
three semicircular canals.

The cochlea is a coiled tube that bears a close resemblance to the shell of a
snail, which is what the word means in Greek. Along its length the cochlea is
divided into three fluid-filled canals: the vestibular canal, the cochlear canal,
and the tympanic canal. The partition between the cochlear canal and the
tympanic canal is called the basilar membrane. Embedded in the basilar
membrane is the spiral-shaped organ of Corti. The sensory cells in the organ of
Corti have thousands of hairlike projections that receive sound vibrations from
the middle ear and send them on to the brain via the auditory nerve. In the
brain they are recognized and interpreted as specific sounds.

3. HEARING

Sound is a series of vibrations moving as waves through air or other gases,

liquids, or solids. A ringing bell, for example, sets off vibrations in the air.
Detection of these vibrations, or sound waves, is called hearing. The detection
of vibrations passing through the ground or water is also called hearing. Some
animals can detect only vibrations passing through the ground, and others can
hear only vibrations passing through water.

168 CHAPTER EIGHT

Humans, however, can hear vibrations passing through gases, solids, and
liquids. Sometimes sound waves are transmitted to the inner ear by a method
of hearing called bone conduction. For example, people hear their own voice
partly by bone conduction. The voice causes the bones of the skull to vibrate,
and these vibrations directly stimulate the sound-sensitive cells of the inner ear.
Only a relatively small part of a normal person’s hearing depends on bone
conduction. If you crunch a biscuit in your mouth, for instance, with your mouth
tightly closed, you can hear the noise made by your chewing act through bone
conduction.

Humans hear primarily by detecting airborne (i.e., carried by air molecules)

sound waves, which are collected by the auricles. The auricles also help locate
the direction of sound. Although some people have auricular muscles so well-
developed that they can wiggle their ears, human auricles, when compared to
those of other mammals, have little importance. Many mammals, especially
those with large ears, such as rabbits, can move their auricles in many
directions so that sound can be picked up more easily.

After being collected by the auricles, sound waves pass through the outer
auditory canal to the eardrum, causing it to vibrate. The vibrations of the
eardrum are then transmitted through the ossicles, the chain of bones in the
middle ear. As the vibrations pass from the relatively large area of the eardrum
through the chain of bones, which have a smaller area, their force is
concentrated. This concentration amplifies, or increases, the sound.

When the sound vibrations reach the stirrup, the stirrup pushes in and out of
the oval window. This movement sets the fluids in the vestibular and tympanic
canals in motion. To relieve the pressure of the moving fluid, the membrane of
the oval window bulges out and in. The alternating changes of pressure in the
fluid of the canals cause the basilar membrane to move. The organ of Corti,
which is part of the basilar membrane, also moves, bending its hairlike

CHAPTER EIGHT 169

projections. The bent projections stimulate the sensory cells to transmit
impulses along the auditory nerve to the brain.

Human ears are capable of perceiving an extraordinarily wide range of

changes in loudness, the tiniest audible sound being about one trillion times
less intense than a sound loud enough to cause the ear pain. The loudness or
intensity of a noise is measured in a unit called the decibel. The softest audible
sound to humans is 0 decibels, while painful sounds are those that rise above
140 decibels. The decibel is a unit of measure (abbreviated dB) originally used
to compare sound intensities and subsequently electrical or electronic power
outputs. Today, it is also used to compare voltages. An increase of 10 dB is
equivalent to a 10-fold increase in intensity or power, and a 20-fold increase in
voltage. A whisper has an intensity of 20 dB. A jet aircraft taking off nearby
(140 dB) is the threshold of pain.

DECIBELS TYPICAL SOUND

0 dB threshold of hearing
10 dB rustle of leaves in gentle breeze
10 dB quiet whisper
20 dB average whisper
20-50 dB quiet conversation
40-45 dB hotel; theater (between performances)
50-65 dB loud conversation
65-70 dB traffic on busy street
65-90 dB train
75-80 dB factory (light/medium work)
90 dB heavy traffic
90-100 dB thunder
110-140 dB jet aircraft at takeoff
130 dB threshold of pain
140-190 dB space rocket at takeoff

170 CHAPTER EIGHT

Besides loudness, the human ear can detect a sound’s pitch, which is related
to a sound’s vibration frequency, or the number of sound waves passing into
the ear in a given period. The greater the frequency, the higher the pitch. The
maximum range of human hearing includes sound frequencies from about 15
to about 18,000 waves, or cycles, per second (i.e., 15 Hz to 18000 Hz).
Because the human ear cannot hear very low frequencies, the sound of one’s
own heartbeat is inaudible. At the other end of the scale, a highly pitched
whistle producing 30,000 cycles per second is not audible to the human ear,
but a dog can hear it.

The third characteristic of sound detected by the human ear is tone. The ability
to recognize tone enables humans to distinguish a violin from a clarinet when
both instruments are playing the same note. The least noticeable change in
tone that can be picked up by the ear varies with pitch and loudness.

Another sonic phenomenon, known as masking, occurs because lower-pitched

sounds tend to deafen the ear to higher-pitched sounds. To overcome the
effects of masking in noisy places, people are forced to raise their voices.

The ear, oddly enough, not only receives sounds but can generate its own
noises, called spontaneous otoacoustic emissions. About 30 percent of all
people experience them. Some hear their ears' sounds, typically a faint
buzzing, if they are in a completely quiet room, but most people do not notice
them at all. In particularly bad cases, people have walked into doctors' offices
complaining of a ringing in their ears, and the doctors could hear the noises
while standing next to the patients. Such sounds, while certainly annoying,
have never been thought to be more than that. But psychologists at the State
University of New York at Buffalo have found that otoacoustic emissions may
actually produce an unusual type of hearing loss, one that occurs without
physical damage to any parts of the ear.

The researchers' evidence comes not from humans but from chinchillas. These
small rodents are prized by some people for their long, silky fur, but to a

CHAPTER EIGHT 171

sensory psychologist they have a different appeal: they have a hearing range
similar to that of humans. Psychologist Salvi and his colleagues were
experimenting with a chinchilla when, with a small microphone, they discovered
a sound coming from the animal's ear. The rodent's ears were producing
otoacoustic emissions at a frequency of 4,000 hertz (vibrations per second).

To see what effect these sounds might have on the chinchilla's hearing, the
researchers placed microelectrodes in the animal's auditory nerve (an
experiment they couldn't have done with humans) to test it for sensitivity to
specific frequencies. They found that the chinchilla was insensitive to external
sounds at frequencies around 4,000 hertz. The constant buzz from its own ears
was apparently causing its brain to ignore that frequency.

Otoacoustic emissions are thought to be produced by delicate microscopic hair

cells in the outer ear. Why they are produced by chinchillas and people, and
why they are produced by some people and not by others, remains largely a
mystery. But Salvi now thinks the sounds create a sort of busy signal in the
auditory nerve. “The ear is the cause of its own hearing loss,” he says. “The
hearing loss is the result of the masking effect only, and there is no cellular or
long-term damage to the ear.” Salvi next hopes to find out whether humans
experience similar types of hearing loss.

4. SPEECH PERCEPTION

Fromkin and Rodman (1988) provide the simplest explanation for speech
perception. They argue that when you hear a car backfire, you may wonder
whether the sound represented a gunshot or a backfiring. Your perception of
the particular acoustic signal and your knowledge of what creates different
sounds result in your assigning some "meaning" to the sounds you heard.
Similarly, when you hear the sounds represented by the phonetic transcription
[ʤæk ɪz ən ˈi:dɪət], you assign the meaning "Jack is an idiot" to the sound
signal. The acoustic signal, however, does not reach our ears in phonemic
segment-sized chunks; it is a semicontinuous signal. In order for us to process

172 CHAPTER EIGHT

it as speech, it must be segmented into phonemes, words, phrases, and
sentences.

Speech perception is a process by which we segment the continuous signal

and, in so doing, may "mischunk" or misperceive the speaker's intended
utterance. The difficulties inherent in speech perception are compounded by
the fact that the speech signal for the "same" utterance varies greatly from
speaker to speaker and from one time to the next by the same speaker.
Nevertheless the brain is able to analyze these different signals, conclude that
they are the same linguistically, segment the utterance into a
phonetic/phonological string of words, "look up" the meaning of these words in
the mental dictionary, technically called the lexicon, analyze the linear string of
words into a hierarchical syntactic structure, and, most of the time, end up with
the intended meaning.

All this work is done so quickly we are unaware that it is going on at all. Despite
the variation between speakers and occurrences, there must be certain
invariant features of speech sounds that permit us to perceive a /d/ or an /ə/
produced by one speaker as identical phonologically with a /d/ and /ə/
produced by another. The relations between the formants (i.e., a frequency
range where vowel sounds are at their most distinctive and characteristic pitch)
of the vowels of one speaker are similar to those of another speaker of the
same language, even though the absolute frequencies may differ.

When a stop consonant is produced, the signal is interrupted slightly, and the
frequency of the "explosion" that occurs at the release of the articulators in
producing stop consonants differs from one consonant to another. The
transitions between consonants and vowels provide important information as to
the identity of the consonants. After voiceless consonants, the onset of vowel
formants starts at higher frequencies than after voiced consonants. Different
places of articulation influence the starting frequencies of formant onsets.

CHAPTER EIGHT 173

There are many such acoustic cues that, together with our knowledge of the
language we are listening to, permit us to perform a "phonetic analysis" on the
incoming acoustic signal. Confusions may also be disambiguated by visual,
lexical, syntactic, and semantic cues. Speech communication often occurs in a
noisy environment, but we can still pick out of the sound signal those aspects
that pertain to speech. We are thus able to ignore large parts of the acoustic
signal in the process of speech perception, which has led to the view that the
human auditory system—perhaps in the course of evolution—has developed a
special ability to detect and process speech cues.

To understand an utterance we must, in some fashion, retrieve information

about the words in that utterance, discover the structural relationship and
semantic properties of those words, and interpret these in the light of the
various pragmatic and discourse constraints operating at the time. Further, all
of this takes place at a remarkably rapid pace. Analyzing the speech signal in
speech perception is a necessary but not sufficient step in understanding a
sentence or utterance. Suppose you heard someone say:

A sniggle blick is procking a slar

and were able to perceive the sounds as

/ə snɪgəl blɪk ɪz prɒkɪn ə slɜ:/

You would still be unable to assign a meaning to the sounds, because the
meaning of a sentence depends on the meanings of its words, and the only
English lexical items in this string are the morphemes a, is, and -ing. The
sentence lacks any English content words. You can only know that the
sentence has no meaning by attempting a lexical lookup of the phonological
strings you construct; finding no entries for sniggle, blick, prock, or slar in your
mental dictionary tells you that the sentence is composed of nonsense strings.
If instead you heard someone say

The cat chased the rat

174 CHAPTER EIGHT

through a lexical lookup process you would conclude that an event concerning
a cat, a rat, and the activity of chasing had occurred. Who chased whom is
determined by syntactic processing. That is, processing speech to get at the
meaning of what is said requires syntactic analysis as well as knowledge of
lexical semantics. Stress and intonation provide some cues to syntactic
structure. We know, for example, that the different meanings of the sentences

He lives in the white house

He lives in the White House

can be signaled by differences in their stress patterns. Relative loudness,

pitch, and duration of syllables thus provide important information in speech
perception.

5. SENSATIONS OF TONE

The ears of an average young person are sensitive to all sounds from about 15
Hz to 20,000 Hz. The hearing of older persons is less acute, particularly to the
higher frequencies. The degree to which a sensitive ear can distinguish
between two pure notes of slightly different loudness or slightly different
frequency varies in different ranges of loudness and frequency. Differences of
about 20 percent in loudness (about one decibel, dB), and 0.33 percent in
frequency (about 1/20 of a note) can be distinguished in sounds of moderate
intensity at the frequencies to which the ear is most sensitive (about 1,000 to
2,000 Hz). In this same range, the difference between the softest sound that
can be heard and the loudest sound that can be distinguished as sound (louder
sounds are “felt,” or perceived, as painful stimuli) is about 120 dB (about one
trillion times as loud).

All of these sensitivity tests refer to pure tones, such as those produced by a
tuning fork. Even for such pure tones the ear is imperfect. Notes of identical
frequency but greatly differing intensity may seem to differ slightly in pitch.
More important is the difference in apparent relative intensities with different

CHAPTER EIGHT 175

frequencies. At high intensities the ear is approximately equally sensitive to
most frequencies, but at low intensities the ear is much more sensitive to the
middle high frequencies than to the lowest and highest. Thus, sound-
reproducing equipment that is functioning perfectly will seem to fail to
reproduce the lowest and highest notes if the volume is decreased.

EXERCISE

1. Ensure that you understand the following terms:

auditory phonetics ossicles organ of Corti

stapes hearing auditory nerve
incus cochlea bone conduction
malleus vestibule loudness
cerumen semicircular canals masking
eustachian tube basilar membrane

2. What is a decibel and how does it relate to sound pressure level and
acoustic intensity?

3. Review the following terms:

hearing
speech perception
sensation of tone
tone
loudness
balance

176 CHAPTER EIGHT

 BIBLIOGRAPHY

Bird, S. & Stegen, O. (1993). Principles of F0 measurement. Poster paper,

Fourth Conference on Laboratory Phonology. Oxford.

Bird, S., & Stegen, O. (1995). The Bamileke Dschang associative construction:
instrumental findings (EUCCS/RP-66). Centre for Cognitive Science,
University of Edinburgh.

Clements, N. G. (1981). Downdrift in a tone language with four tone levels: a

report on some preliminary measurements. Unpublished ms. (Published
in York Papers in Linguistics, Vol. 15 1991)

Clements, G. N. (1990). The status of register in intonation: comments on the

paper by Ladd and by Inkelas and Leben. In J. Kingston & M. Beckman
(Ed.), Papers in Laboratory Phonology I: Between the grammar and
physics of speech (pp. 58-71). Cambridge: Cambridge University Press.

Connell, B. A. & Ladd, D. R. (1990). Aspects of Pitch Realization in Yoruba.

Phonology, 7 1, 1-29.

Falk, J. S. (1978). Linguistics and language: A survey of basic concepts and

implications. New York: John Wiley & Sons.

Fromkin, V., Rodman, R., & Hyams, N. (2003). An introduction to language.

United States: Thomson.

Gandour, J., Potisuk, S. & Dechongkit, S. (1994). Tonal coarticulation in Thai.

Journal of Phonetics, 22 4, 477-492.

BIBLIOGRAPHY 177
Gussenhoven, C. & Rietveld, A. (1988). Fundamental frequency in Dutch:
Testing three hypotheses. Journal of Phonetics, 16 355-369.

Hombert, J.-M. (1974). Universals of Downdrift: their phonetic basis and

significance for a theory of tone. Studies in African Linguistics, Suppl. 5
169-183.

Hombert, J.-M. (1977). Consonant types, vowel height, and tone in Yoruba.
Studies in African Linguistics, 8 2, 173-190.

Hombert, J.-M. (1978). Consonant types, vowel quality, and tone. In F. V. (Ed.),
Tone: a linguistic survey (pp. 77-111). New York: Academic Press.

Katamba, F. (1988). An introduction to phonology. London: Longman Group

UK Limited.

Ladd, D. R. (1984). Declination: a review and some issues. Phonology

Yearbook, 1 53-74.

Ladd, D. R. (1990). Metrical representation of pitch register. In J. Kingston & M.

Beckman (Ed.), Papers in Laboratory Phonology I: Between the
grammar and physics of speech Cambridge: Cambridge University
Press.

Ladd, D. R. (1992). An introduction to intonational phonology. In G. J. Docherty

& D. R. Ladd (Ed.), Papers in Laboratory Phonology II: Gesture,
segment, prosody. Cambridge: Cambridge University Press.

Ladd, D. R. & Johnson, C. (1987). 'Metrical' factors in the scaling of sentence-

initial accent peaks. Phonetica, 44 238-245

Ladefoged, P. (1982). A course in phonetics. New York: Harcourt, Brace

Jovanovich.

178 BIBLIOGRAPHY
Laniran, Y. (1992a). Intonation in a tone language: the phonetic implentation of
tone in Yoruba. Ph.D., Cornell,

Laniran, Y. (1992b). Phonetic aspects of tone realization in Igbo. Progress

Reports from Oxford Phonetics, 5 35-51.

Laver, J. (1994). Principles of Phonetics. Cambridge: Cambridge University

Press.

Leben, W. R. (1996). Prosodic domains and the theory of tone. Paper

presented to the 27th ACAL, Gainesville, FL.

Liberman, M. & Pierrehumbert, J. (1984). Intonational invariance under

changes in pitch range and length. In M. Aronoff & R. Oehrle (Ed.),
Language and Sound Structure (pp. 157-233). Cambridge MA: MIT.

Liberman, M., Schultz, J. M., Hong, S. & Okeke, V. (1993). The phonetic
interpretation of tone in Igbo. Phonetica, 50 147-160.

Lieberman, P., & Blumstein, S. E. (1988). Speech physiology, speech

perception, and acoustic phonetics. Cambridge: Cambridge University
Press.

Lindau, M. (1986). Testing a model of intonation in a tone language. Journal of

the Acoustical Society of America, 80 757-764

Microsoft Corporation. (2004). Microsoft Encarta reference library [Computer

Software]. Redmond, WA.: Microsoft Corporation.

Ohala, J. J. (1978). The production of tone. In V. Fromkin (Ed.), Tone: a

linguistic survey (pp. 5-39). New York: Academic Press.

BIBLIOGRAPHY 179
Perrin, M. J. (1974). Mambila. In J. Bendor-Samuel (Ed.), Ten Nigerian Tone
Systems (pp. 93-108). Jos: Institute of Linguistics.

Perrin, M. (n.d.). The tone system in Mambila: some further comments.

Pierrehumbert, J. & Beckman, M. (1988). Japanese Tone Structure. Cambridge

MA: MIT Press.

Snider, K. & van der Hulst, H. (1993). Issues in the representation of tonal
register. In H. van der Hulst & K. Snider (Ed.), The Phonlogy of Tone
(pp. 1-27). Berlin: Mouton de Gruyter.

Stewart, J. M. (1983). Downstep and Floating Low Tones in Adioukrou. Journal

of African Languages and Linguistics, 5 57-78.

Welmers, W. (1973). African Language Structures. Berkeley: University of

California Press.

Zee, E. (1978). The interaction of tone and vowel quality. UCLA Working
Papers in Phonetics, 41 53-67.

180 BIBLIOGRAPHY